Compare / Claude Haiku 4.5 vs GPT-6.1 Terra
Claude Haiku 4.5 vs GPT-6.1 Terra
Which is cheaper, which is more capable, and which does better at each kind of work, from 96 results on 6 sources that measured both models.
Of the 96 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.
Results as of 19 November 2026 (catalogue release 2026-11-19-3608133e8331). See also Anthropic vs OpenAI, every model from both labs.
Claude Haiku 4.5 or GPT-6.1 Terra? The short answer
Pick GPT-6.1 Terra: it scores higher (Spring Prompt overall 77 against 7) and costs less.
- Spring Prompt overall: Claude Haiku 4.5 7, GPT-6.1 Terra 77 (out of 100; how the score works).
- GPT-6.1 Terra is cheaper to run: $0.004 against $0.006 per product listing on CatalogBench (42% less).
- GPT-6.1 Terra is newer: released 26 Oct 2026, against 15 Oct 2025 for Claude Haiku 4.5.
- On our own benchmarks, on the results both models have: GPT-6.1 Terra does better for product listings, decks from an analysis, user surveys and marketing planning.
- On third-party benchmarks, on the results both models have: GPT-6.1 Terra does better for agents and tool use and reasoning and knowledge.
Head to head
Every fact side by side
| Claude Haiku 4.5 | GPT-6.1 Terra | |
|---|---|---|
| Price per million tokens | $1.00 in · $5.00 out | $1.00 in · $5.00 out |
| Blended (3 in : 1 out) | $2.00 | $2.00 |
| Cheapest host, blended | $2.00 (Amazon Bedrock) | $2.00 (Azure) |
| Cost per product listing (CatalogBench) | $0.006 | $0.004 |
| Cached input, per million | $0.10 cached input | $0.050 cached input |
| Intelligence Index | 16.9 | – |
| Spring Prompt overall | 7 | 77 |
| Output speed, tokens a second | – | – |
| Context window | 200,000 tokens | 1,050,000 tokens |
| Released | 15 Oct 2025 | 26 Oct 2026 (newer) |
| Weights | Closed | Closed |
| Developer based in | the US | the US |
Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.
Which is better for what
Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.
Their slides, side by side
The same brief for both: Manchester store investment review for Hearthside Coffee, an invented company. First slide of each deck from DeckBench.
From our research
- Claude Haiku 5.5 Is Far Better Than Haiku 4.5 at a Tenth of the Price. GPT-6 Luna Still Beats It at Business Work. 8 Oct 2026
- GPT-6.1 Terra review: close to the top of our business tests at half the price of Sol 5 Nov 2026
- No Benchmark Ranked This Month's AI Models the Way Reviewers Did. Ours Included. 19 Nov 2026
- GPT-6.2 review: Sol writes the better listings and decks, Astra the better analysis, at four times the price 18 Nov 2026
Every shared result
Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.
| Measure | Claude Haiku 4.5 | GPT-6.1 Terra |
|---|---|---|
| BulletBench measured by Spring Prompt · 3 results | ||
| Cost per game · Blitz 3+2US dollars, lower is better 6 · 9 of 23 measured | $0.0318 | $0.0706 |
| Cost per game · Bullet 60sUS dollars, lower is better 45 · 47 of 62 measured | $0.0163 | $0.0172low reasoning |
| Ladder Elo · Bullet 60sladder Elo, higher is better 41 · 46 of 62 measured | 445 | 358low reasoning |
| CatalogBench measured by Spring Prompt · 3 results | ||
| Cost per productUS dollars, lower is better 12 · 9 of 36 measured | $0.0061 | $0.0036 |
| Reliably publish-ready% of products, higher is better 32 · 9 of 35 measured | 1.8% | 64.3% |
| Reliably publish-ready · sales brief% of products, higher is better 30 · 4 of 35 measured | 0.0% | 67.9% |
| DeckBench measured by Spring Prompt · 3 results | ||
| Cost per deckUS dollars, lower is better 12 · 16 of 36 measured | $0.0461 | $0.0672 |
| Deck ratingrating, higher is betterThe rating weighs each win by how strong the other deck was; the win rate counts wins against whichever decks a model happened to meet, so the two can disagree. 32 · 9 of 36 measured | 646 | 1,280 |
| Accurate decks% of tasks, higher is better 27 · 1 of 36 measured | 16.7% | 100.0% |
| ROASBench measured by Spring Prompt · 3 results | ||
| Overall scorescore out of 100, higher is better 35 · 2 of 36 measured | 17.5 | 58.4 |
| Audience scorescore out of 100, higher is better 35 · 1 of 36 measured | 30.9 | 69.4 |
| Business scorescore out of 100, higher is better 35 · 3 of 36 measured | 7.3 | 59.1 |
| SurveyBench measured by Spring Prompt · 3 results | ||
| Analysis ratingrating, higher is betterThe rating weighs each win by how strong the other analysis was; the win rate counts wins against whichever analyses a model happened to meet, so the two can disagree. 26 · 10 of 27 measured | 561 | 1,217 |
| Cost per taskUS dollars, lower is better 10 · 9 of 27 measured | $0.0635 | $0.0569 |
| Analysis scorepoints of 100, higher is better 20 · 6 of 27 measured | 56.8 | 91.6 |
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Humanity's Last Exam% of questions, higher is better 157 · 21 of 250 measured | 10.4%thinking reasoning | 46.2%max reasoning |
| Terminal-Bench 4.0% of tasks, higher is better 72 · 7 of 120 measured | 0.0%thinking reasoning | 55.6%max reasoning |
| Long-context reasoning (AA-LCR)% of questions, higher is better 89 · 20 of 241 measured | 74.3%thinking reasoning | 82.7%max reasoning |
Show all 96 results (78 more)
| Measure | Claude Haiku 4.5 | GPT-6.1 Terra |
|---|---|---|
| BulletBench measured by Spring Prompt · 12 results | ||
| Ladder Elo · Lightning 10+1ladder Elo, higher is better 42 · 32 of 45 measured | 33 | 491low reasoning |
| Cost per game · Lightning 10+1US dollars, lower is better 39 · 34 of 45 measured | $0.0095 | $0.0064low reasoning |
| Games lost on time · Blitz 3+2% of games, lower is better 1 · 10 of 23 measured | 0.0% | 25.0% |
| Games lost on time · Bullet 60s% of games, lower is better 31 · 52 of 62 measured | 25.0% | 75.0%low reasoning |
| Games lost on time · Lightning 10+1% of games, lower is better 42 · 37 of 45 measured | 83.3% | 62.5%low reasoning |
| Invalid moves · Blitz 3+2% of moves, lower is better 23 · 1 of 23 measured | 0.9% | 0.0% |
| Invalid moves · Bullet 60s% of moves, lower is better 56 · 1 of 62 measured | 0.3% | 0.0%low reasoning |
| Invalid moves · Lightning 10+1% of moves, lower is better 1 · 1 of 45 measured | 0.0% | 0.0%low reasoning |
| Ladder Elo · Blitz 3+2ladder Elo, higher is better 20 · 12 of 23 measured | 491 | 683 |
| Median move time · Blitz 3+2milliseconds, lower is better 4 · 8 of 23 measured | 1.3 s | 3.7 s |
| Median move time · Bullet 60smilliseconds, lower is better 36 · 53 of 62 measured | 1.2 s | 2.8 slow reasoning |
| Median move time · Lightning 10+1milliseconds, lower is better 41 · 40 of 45 measured | 1.3 s | 1.3 slow reasoning |
| CatalogBench measured by Spring Prompt · 28 results | ||
| Channel compliance% of products, higher is better 1 · 35 of 36 measured | 100.0% | 98.2% |
| Channel rules broken% of products, lower is better 1 · 33 of 36 measured | 0.0% | 1.8% |
| Channel rules broken · sales brief% of products, lower is better 36 · 1 of 36 measured | 4.2% | 0.0% |
| Claims to check% of products, lower is better 34 · 1 of 34 measured | 22.6% | 0.0% |
| Claims to check · sales brief% of products, lower is better 34 · 1 of 34 measured | 44.6% | 0.0% |
| Conflicts caught% of conflicts, higher is better 1 · 1 of 36 measured | 100.0% | 100.0% |
| Content quality% of checks, higher is better 33 · 13 of 36 measured | 94.4% | 97.7% |
| Content quality · sales brief% of checks, higher is better 34 · 10 of 36 measured | 93.9% | 98.3% |
| Decision accuracy% of decisions, higher is better 35 · 8 of 36 measured | 90.6% | 98.1% |
| Failed outputs% of products, lower is better 1 · 1 of 36 measured | 0.0% | 0.0% |
| Failed outputs · sales brief% of products, lower is better 1 · 1 of 36 measured | 0.0% | 0.0% |
| Field accuracy% of missing fields, higher is better 18 · 10 of 36 measured | 93.7% | 94.4% |
| Invented values% of filled values, lower is better 36 · 16 of 36 measured | 6.2% | 1.5% |
| Missing UK information% of products, lower is better 1 · 1 of 36 measured | 0.0% | 0.0% |
| Missing UK information · sales brief% of products, lower is better 1 · 1 of 36 measured | 0.0% | 0.0% |
| Not findable% of products, lower is better 31 · 33 of 36 measured | 8.9% | 10.7% |
| Not findable · sales brief% of products, lower is better 27 · 35 of 36 measured | 6.6% | 13.7% |
| Publish-ready listings% of products, higher is better 33 · 10 of 36 measured | 12.5% | 74.4% |
| Publish-ready listings · sales brief% of products, higher is better 33 · 5 of 36 measured | 3.0% | 74.4% |
| Time per productseconds, lower is better 6 · 12 of 36 measured | 8.10 | 10.1 |
| Unsupported claimsclaims per product, lower is better 36 · 6 of 36 measured | 2.69 | 0.03 |
| Unsupported claims% of products, lower is better 36 · 3 of 36 measured | 73.2% | 0.6% |
| Unsupported claims · sales briefclaims per product, lower is better 36 · 5 of 36 measured | 5.32 | 0.09 |
| Unsupported claims · sales brief% of products, lower is better 36 · 2 of 36 measured | 91.7% | 1.8% |
| Wrong attributes% of products, lower is better 36 · 12 of 36 measured | 48.2% | 14.9% |
| Wrong attributes · sales brief% of products, lower is better 36 · 11 of 36 measured | 51.2% | 14.9% |
| Wrong category or variant% of products, lower is better 34 · 7 of 36 measured | 7.7% | 1.8% |
| Wrong category or variant · sales brief% of products, lower is better 34 · 7 of 36 measured | 6.6% | 1.8% |
| DeckBench measured by Spring Prompt · 14 results | ||
| Caveat dropped% of tasks, lower is better 1 · 1 of 36 measured | 0.0% | 0.0% |
| Clean layout% of tasks, higher is better 16 · 5 of 36 measured | 50.0% | 83.3% |
| Design quality% of the maximum, higher is better 32 · 8 of 36 measured | 47.6% | 80.8% |
| Draft figure quoted% of tasks, lower is better 29 · 1 of 36 measured | 50.0% | 0.0% |
| Findings missing% of tasks, lower is better 25 · 1 of 36 measured | 16.7% | 0.0% |
| Head-to-head win rate% of comparisons, higher is betterThe rating weighs each win by how strong the other deck was; the win rate counts wins against whichever decks a model happened to meet, so the two can disagree. 32 · 26 of 36 measured | 29.8% | 41.9% |
| Layout defects% of tasks, lower is better 16 · 5 of 36 measured | 50.0% | 16.7% |
| Misleading metric used% of tasks, lower is better 1 · 1 of 36 measured | 0.0% | 0.0% |
| Presentable decks% of tasks, higher is better 17 · 5 of 36 measured | 0.0% | 50.0% |
| Recommendation late or wrong% of tasks, lower is better 28 · 1 of 36 measured | 50.0% | 0.0% |
| Slides needing work% of tasks, lower is better 26 · 12 of 36 measured | 66.7% | 33.3% |
| Time per deckseconds, lower is better 6 · 19 of 36 measured | 51.7 | 135 |
| Unsupported claims% of tasks, lower is better 26 · 1 of 36 measured | 33.3% | 0.0% |
| Unsupported numbers% of tasks, lower is better 1 · 1 of 36 measured | 0.0% | 0.0% |
| ROASBench measured by Spring Prompt · 6 results | ||
| Consistency scorescore out of 100, higher is better 35 · 1 of 36 measured | 8.7 | 48.8 |
| Contribution profitsimulated US dollars, higher is better 36 · 10 of 36 measured | −$205,001 | $701,059 |
| Cost of a runUS dollars, lower is better 5 · 11 of 36 measured | $0.0298 | $0.11 |
| Months over budgetmonths of 12, lower is better 34 · 1 of 36 measured | 2 | 0 |
| Planning scorescore out of 100, higher is better 27 · 10 of 36 measured | 54.4 | 55.0 |
| Return on ad spendprofit per $1 spent, higher is better 36 · 3 of 36 measured | −$0.21 | $0.71 |
| SurveyBench measured by Spring Prompt · 17 results | ||
| Head-to-head win rate% of comparisons, higher is betterThe rating weighs each win by how strong the other analysis was; the win rate counts wins against whichever analyses a model happened to meet, so the two can disagree. 26 · 11 of 27 measured | 24.7% | 56.3% |
| Leading or double-barrelled questions% of tasks, lower is better 24 · 5 of 27 measured | 66.7% | 11.1% |
| Mistaken next steps% of tasks, lower is better 22 · 5 of 27 measured | 55.6% | 11.1% |
| Numbers right% of answers, higher is better 26 · 1 of 27 measured | 63.7% | 100.0% |
| Objectives unanswerable% of tasks, lower is better 24 · 18 of 27 measured | 22.2% | 11.1% |
| Precision ignored% of tasks, lower is better 25 · 1 of 27 measured | 100.0% | 0.0% |
| Research scorepoints of 100, higher is better 20 · 6 of 27 measured | 64.9 | 92.2 |
| Sound research% of tasks, higher is better 17 · 6 of 27 measured | 0.0% | 33.3% |
| Stakeholder's view presumed% of tasks, lower is better 18 · 12 of 27 measured | 22.2% | 11.1% |
| Survey plan scorepoints of 100, higher is better 22 · 10 of 27 measured | 83.6 | 93.6 |
| Survey structure% of tasks, lower is better 1 · 12 of 27 measured | 0.0% | 22.2% |
| Time per taskseconds, lower is better 8 · 7 of 27 measured | 92.6 | 92.1 |
| Traps handled% of traps, higher is better 24 · 11 of 27 measured | 62.3% | 94.2% |
| Traps missed% of tasks, lower is better 19 · 8 of 27 measured | 88.9% | 33.3% |
| Unsupported findings% of tasks, lower is better 24 · 1 of 27 measured | 100.0% | 0.0% |
| Wrong numbers% of tasks, lower is better 25 · 1 of 27 measured | 100.0% | 0.0% |
| Wrong recommendation% of tasks, lower is better 19 · 5 of 27 measured | 44.4% | 11.1% |
| Artificial Analysis reported by Artificial Analysis · 1 result | ||
| SciCode% of problems, higher is better 81 · 27 of 121 measured | 42.2%thinking reasoning | 55.1%max reasoning |
Compare them with other models
Quick answers
Is Claude Haiku 4.5 better than GPT-6.1 Terra?
Pick GPT-6.1 Terra: it scores higher (Spring Prompt overall 77 against 7) and costs less.
Which is cheaper, Claude Haiku 4.5 or GPT-6.1 Terra?
GPT-6.1 Terra is cheaper to run: $0.004 against $0.006 per product listing on CatalogBench (42% less).
Which is better for product listings, Claude Haiku 4.5 or GPT-6.1 Terra?
GPT-6.1 Terra ranks 6 of 35 models on sale for turning a sparse product feed and photos into listings that can go live, against 32 for Claude Haiku 4.5.
Which is better for decks from an analysis, Claude Haiku 4.5 or GPT-6.1 Terra?
GPT-6.1 Terra ranks 9 of 35 models on sale for turning a finished analysis into a deck you could present as it is, against 31 for Claude Haiku 4.5.
Which is better for user surveys, Claude Haiku 4.5 or GPT-6.1 Terra?
GPT-6.1 Terra ranks 10 of 26 models on sale for planning a user survey and reading its results without being misled, against 25 for Claude Haiku 4.5.
Which is better for marketing planning, Claude Haiku 4.5 or GPT-6.1 Terra?
GPT-6.1 Terra ranks 2 of 35 models on sale for planning a year of ad spend without overspending, against 34 for Claude Haiku 4.5.
Which is better for agents and tool use, Claude Haiku 4.5 or GPT-6.1 Terra?
GPT-6.1 Terra ranks 7 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 98 for Claude Haiku 4.5.
Which has the bigger context window, Claude Haiku 4.5 or GPT-6.1 Terra?
GPT-6.1 Terra: 1,050,000 tokens, against 200,000 for Claude Haiku 4.5.
Run Claude Haiku 4.5 and GPT-6.1 Terra on your own prompt.
Benchmarks aren't your task. Test both side by side on your prompt, with every answer checked against your rules and the cost of each one.

