GPT-6.1 Sol: benchmarks, price and rank
GPT-6.1 Sol from OpenAI is 8th of 123 rated AI models on Spezia's leaderboard, strongest on FrontierMath and LMArena WebDev. It costs $2 in, $10 out per million tokens.
- 8th of 123 rated models
- Rating 1,852
- First on 1 of 11 benchmarks
- Closed
- Released Sep 29, 2026
- 1.1 million tokens of context
- Knows the world to April 2026
Where it is strongest
- 2nd of 76 on FrontierMath
- 6th of 98 on LMArena WebDev
- 5th of 78 on SimpleQA Verified
- 5th of 60 on Vending-Bench 2
- 7th of 68 on SimpleBench
Where it falls behind
In the top half on every benchmark it has a result on.
Price: $2 in, $10 out per million tokens.
Its headline results
The benchmarks that count most in the rating, with its place among the models measured the same way, and where each number came from.
- APEX-Agents60%
Long tasks written by investment banking analysts, management consultants and corporate lawyers from their own work, done across files and apps in a realistic work setup.
19th of 56. Independent result from LLM Stats.
- Vending-Bench 2$12,861
The model runs a simulated vending machine business for a year on its own: finding suppliers, ordering stock, setting prices and paying the daily fees.
5th of 60. Independent result from Epoch AI.
- DeepSWE 1.175.2%
Software engineering tasks run in the same simple agent setup for every model.
5th of 27. The maker's own number, from OpenAI, via LLM Stats.
- SimpleBench82.9%
Multiple-choice puzzles about everyday situations, space, time and people, written to trip up models that pattern-match instead of think. Ordinary adults do well on them.
7th of 68. Independent result from Epoch AI.
- SimpleQA Verified73.9%
Short factual questions with one checked answer, asked without web search.
5th of 78. Independent result from Epoch AI.
- FrontierMath93.7%
Hundreds of new, unpublished math problems written by mathematicians, from hard undergraduate level up to research, with answers a computer can check.
2nd of 76. Independent result from Epoch AI.
- MMMU-Pro86.0%
College exam questions that need a picture to answer, such as charts, diagrams, maps and chemical structures, with more answer options than the original MMMU and some questions shown only as a screenshot.
Independent result from LLM Stats.
- LMArena text1,446
People put a question to two anonymous models and vote for the better answer.
43rd of 107. Independent result from LMArena.
- LMArena WebDev1,743
People ask two anonymous models to build the same web page or app and vote for the better one.
6th of 98. Independent result from LMArena.
- HealthBench Professional64.2%
Conversations in the style of real clinicians' chats with AI, graded against rubrics written by physicians.
6th of 19. The maker's own number, from OpenAI, via LLM Stats.
6 more results in its full profile.
Every result with its source, what it costs per task, what jumped with its release, side by side with any other model, and answers to your questions about when to use it.
Models ranked near it
How the rating works, and every benchmark it uses: the model leaderboard.