{"licence": "Creative Commons Attribution 4.0 International (CC BY 4.0)", "licence_url": "https://creativecommons.org/licenses/by/4.0/", "attribution": "Spring Prompt (springprompt.com)", "source_url": "https://springprompt.com/benchmarks/deckbench", "release": "2026-11-19-3608133e8331", "test_your_prompt": "These are our tasks, not yours. Test your own prompt on these models: https://springprompt.com/prompt-test", "results": [{"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 1563.4, "ci_low": 1477.8, "ci_high": 1692.1, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 1551.3, "ci_low": 1433.6, "ci_high": 1729.5, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 1437.8, "ci_low": 1370.3, "ci_high": 1544.9, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 1403.1, "ci_low": 1311.4, "ci_high": 1511.6, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 1375.1, "ci_low": 1304.9, "ci_high": 1463.2, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 1370.0, "ci_low": 1276.9, "ci_high": 1493.4, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 1333.6, "ci_low": 1249.2, "ci_high": 1445.1, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 1322.9, "ci_low": 1245.0, "ci_high": 1405.8, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 1280.2, "ci_low": 1195.9, "ci_high": 1389.8, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 1266.0, "ci_low": 1176.5, "ci_high": 1366.9, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 1256.1, "ci_low": 1190.5, "ci_high": 1332.9, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 1252.4, "ci_low": 1193.5, "ci_high": 1332.5, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 1241.7, "ci_low": 1173.4, "ci_high": 1325.8, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 1223.2, "ci_low": 1095.4, "ci_high": 1329.1, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 1203.8, "ci_low": 1146.9, "ci_high": 1259.5, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 1194.4, "ci_low": 1147.5, "ci_high": 1276.7, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 1187.9, "ci_low": 1081.9, "ci_high": 1318.2, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 1142.7, "ci_low": 1065.8, "ci_high": 1226.2, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 1120.1, "ci_low": 1050.0, "ci_high": 1187.7, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 1086.0, "ci_low": 983.0, "ci_high": 1185.4, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 1078.8, "ci_low": 983.4, "ci_high": 1175.5, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 1046.2, "ci_low": 943.6, "ci_high": 1120.4, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 1042.6, "ci_low": 950.4, "ci_high": 1135.4, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 1025.3, "ci_low": 953.9, "ci_high": 1101.5, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 1013.2, "ci_low": 879.2, "ci_high": 1112.5, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 991.9, "ci_low": 851.0, "ci_high": 1114.3, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 951.3, "ci_low": 882.5, "ci_high": 1024.6, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 809.3, "ci_low": 677.9, "ci_high": 919.0, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 806.7, "ci_low": 747.6, "ci_high": 853.9, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 771.4, "ci_low": 706.4, "ci_high": 824.5, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 726.2, "ci_low": 632.0, "ci_high": 814.4, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 715.6, "ci_low": 643.5, "ci_high": 787.3, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 645.7, "ci_low": 541.4, "ci_high": 737.7, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 605.2, "ci_low": 494.1, "ci_high": 694.9, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 594.8, "ci_low": 472.3, "ci_high": 692.5, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 585.1, "ci_low": 473.0, "ci_high": 661.9, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 530.0, "ci_low": 390.6, "ci_high": 611.2, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 95.28, "ci_low": null, "ci_high": null, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 88.51, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 87.9, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 86.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 83.89, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 83.38, "ci_low": null, "ci_high": null, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 81.38, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 80.75, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 78.47, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 78.1, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 77.83, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 76.44, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 71.96, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 71.39, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 68.47, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 67.61, "ci_low": null, "ci_high": null, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 67.11, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 66.95, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 66.91, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 66.62, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 61.71, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 61.06, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 60.22, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 59.77, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 58.49, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 56.65, "ci_low": null, "ci_high": null, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 56.05, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 55.2, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 54.37, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 51.39, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 50.81, "ci_low": null, "ci_high": null, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 47.62, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 46.23, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 43.65, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 40.87, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 30.51, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 29.3, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 89.74, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 85.37, "ci_low": null, "ci_high": null, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 84.85, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 82.48, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 79.73, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 78.57, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 76.13, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 75.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 70.88, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 70.83, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 69.77, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 69.74, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 64.91, "ci_low": null, "ci_high": null, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 63.89, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 62.07, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 60.84, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 60.63, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 59.65, "ci_low": null, "ci_high": null, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 51.37, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 48.48, "ci_low": null, "ci_high": null, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 47.46, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 46.01, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 45.45, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 44.59, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 43.64, "ci_low": null, "ci_high": null, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 42.47, "ci_low": null, "ci_high": null, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 41.89, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 41.18, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 39.39, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 37.84, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 35.24, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 31.18, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 29.81, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 28.66, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 25.23, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 24.69, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 24.43, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 0.00857, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 0.00865, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 0.00905, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.01042, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 0.01122, "ci_low": null, "ci_high": null, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 0.01162, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 0.01262, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 0.01477, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 0.02552, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 0.0359, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 0.0448, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 0.04607, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 0.04626, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 0.0465, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 0.06134, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 0.06722, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 0.07745, "ci_low": null, "ci_high": null, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 0.11468, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 0.1355, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 0.14527, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 0.15273, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 0.15295, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 0.16376, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 0.17891, "ci_low": null, "ci_high": null, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 0.18211, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 0.19567, "ci_low": null, "ci_high": null, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 0.20966, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 0.2498, "ci_low": null, "ci_high": null, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 0.25243, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 0.26102, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 0.4537, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.54068, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 0.77156, "ci_low": null, "ci_high": null, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 0.80661, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 1.09215, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 1.39571, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 15.2, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 25.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 28.6, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 30.2, "ci_low": null, "ci_high": null, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 44.5, "ci_low": null, "ci_high": null, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 51.7, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 62.1, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 67.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 73.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 78.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 81.8, "ci_low": null, "ci_high": null, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 83.3, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 83.4, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 89.4, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 93.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 98.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 100.6, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 114.2, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 134.9, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 154.4, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 157.9, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 169.1, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 174.8, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 178.9, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 202.1, "ci_low": null, "ci_high": null, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 217.3, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 228.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 285.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 311.4, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 330.1, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 332.3, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 374.8, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 424.2, "ci_low": null, "ci_high": null, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 437.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 488.6, "ci_low": null, "ci_high": null, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 543.1, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 1288.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 4453.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 7306.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 12086.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 15626.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 16952.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 21019.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 22094.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 27796.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 31368.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 32448.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 33285.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 33765.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 36572.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 45141.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 48462.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 51739.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-astra", "model_name": "GPT-6.2 Astra", "developer": "OpenAI", "configuration": "GPT-6.2 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-18", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-2-sol", "model_name": "GPT-6.2 Sol", "developer": "OpenAI", "configuration": "GPT-6.2 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-17", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-hydrogen", "model_name": "Gemini 4 Hydrogen", "developer": "Google", "configuration": "Gemini 4 Hydrogen", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen4-max", "model_name": "Qwen4-Max", "developer": "Alibaba", "configuration": "Qwen4-Max", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-16", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-ember", "model_name": "Muse Ember", "developer": "Meta", "configuration": "Muse Ember", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-09", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-carbon", "model_name": "Gemini 4 Carbon", "developer": "Google", "configuration": "Gemini 4 Carbon", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-neon", "model_name": "Gemini 4 Neon", "developer": "Google", "configuration": "Gemini 4 Neon", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}]}