Models

541 models with published results from our own benchmarks and licensed sources. Open a model to see every result, with its source and date.

ModelProviderSourcesResults
ling-1tAnt Group13
ling-flash-2.0Ant Group16
llama-13bMeta14
llama-2-13b-chatMeta16
llama-2-7b-chatMeta16
llama-3-70b-instructMeta16
llama-3-8b-instructMeta16
llama-3.1-405b-instructMeta13
llama-3.1-405b-instruct-bf16Meta16
llama-3.1-405b-instruct-fp8Meta16
llama-3.1-nemotron-51b-instructNVIDIA16
llama-3.1-nemotron-70b-instructNVIDIA16
llama-3.1-tulu-3-70bAi215
llama-3.1-tulu-3-8bAi215
llama-3.3-70b-instruct-turboMeta12
llama-3.3-8b-instructAllura Forge13
llama-3.3-nemotron-49b-super-v1NVIDIA15
llama2-70b-steerlm-chatNVIDIA15
longcat-flash-chatMeituan16
longcat-flash-chat-2602-expMeituan17
magistral-medium-2506Mistral16
magistral-small-2509Mistral16
medgemma-27b-text-itGoogle13
mercuryInception16
mimo-v2-omniXiaomi17
mimo-v2-proXiaomi17
minicpm3-4bOpenbmb116
minimax-m2.1-previewMiniMax17
ministral-3-14b-reasoning-2512Mistral16
ministral-3-8b-reasoning-2512Mistral19
ministral-3b-2410Mistral12
mistral-7b-instructMistral16
mistral-7b-instruct-v0.2Mistral16
mistral-large-2402Mistral16
mistral-mediumMistral16
mistral-small-instruct-2409Mistral13
molmo-2-8bAi212
mpt-30b-chatMosaicml15
mpt-7b-chatMosaicml15
muse-glimmerMeta17
muse-sparkMeta18
nanbeige3.5-pro-thinkingNanbeige18
nanbeige4-3b-thinking-2510Nanbeige13
nemotron-4-340b-instructNVIDIA16
nous-hermes-2-mixtral-8x7b-dpoNousresearch15
nvidia-llama-3.3-nemotron-super-49b-v1.5NVIDIA16
nvidia-nemotron-3-ultra-550b-a55b-nvfp4NVIDIA17
nvidia-nemotron-3.5-lightning-30b-a3b-nvfp4NVIDIA17
nvidia-nemotron-nano-12b-v2-vl-bf16NVIDIA16
o1-miniOpenAI16

Want two models side by side? Compare models.