GPT-6 Sol: benchmarks, price and rank
GPT-6 Sol from OpenAI is 21st of 123 rated AI models on Spezia's leaderboard, strongest on Vending-Bench 2 and FrontierMath. It costs $2 in, $10 out per million tokens.
- 21st of 123 rated models
- Rating 1,589
- Rated on 11 benchmarks
- Closed
- Released Sep 22, 2026
- 1.1 million tokens of context
- Knows the world to April 2026
Where it is strongest
- 3rd of 60 on Vending-Bench 2
- 8th of 76 on FrontierMath
- 10th of 103 on GPQA Diamond
Where it falls behind
- 85th of 107 on LMArena text
- 14th of 27 on DeepSWE 1.1
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-Agents54.3%
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.
28th of 56. Independent result from Epoch AI.
- Vending-Bench 2$14,428
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.
3rd of 60. Independent result from Epoch AI.
- DeepSWE 1.168.8%
Software engineering tasks run in the same simple agent setup for every model.
14th of 27. The maker's own number, from OpenAI, via LLM Stats.
- SimpleBench73.1%
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.
20th of 68. Independent result from Epoch AI.
- SimpleQA Verified60.7%
Short factual questions with one checked answer, asked without web search.
18th of 78. Independent result from Epoch AI.
- FrontierMath89.8%
Hundreds of new, unpublished math problems written by mathematicians, from hard undergraduate level up to research, with answers a computer can check.
8th of 76. Independent result from Epoch AI.
- LMArena text1,397
People put a question to two anonymous models and vote for the better answer.
85th of 107. Independent result from LMArena.
- LMArena WebDev1,673
People ask two anonymous models to build the same web page or app and vote for the better one.
14th of 98. Independent result from LMArena.
- GPQA Diamond94.3%
Graduate-level questions in biology, physics and chemistry, written by PhDs so that searching the web does not give the answer away.
10th of 103. Independent result from Epoch AI.
- ARC-AGI-289.6%
Visual puzzles of colored grids, each with a new rule to work out from a few examples.
10th of 69. Independent result from Epoch AI.
2 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.