ArXivLean
Models prove or disprove theorems from recent arXiv papers, stated in Lean 4. A strict comparator checks each submitted proof against the fixed reference statement.
The first set held 41 theorems from March 2026 papers, and the best systems proved about 17 percent. Later sets came in June and September 2026.
- Released
- April 2026
- Kind
- Formal proofs
- Level
- Research frontier
- Scored
- Checked automatically
- Answers
- Written proof
- Tasks
- All public
- Strongest model tested
- GPT-6 Astra #3 on the capability index
- Last in a system card
- July 2026
- Last model added
- October 2026
- Scores refreshed
- November 12th, 2026
Claude Opus 5.5 leads with 47.2%
Best score per model from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best.
Claude Opus 5.5 leads with 47.2%, claims included
Best score per model from every source, including labs’ own numbers and claims Epoch AI collected from reports and papers. Labs pick their own settings, so a claim can beat an independent run of the same model.
- 1 Claude Opus 5.5 high 47.2%
- 2
Gemini 4 Argon high 39.6% - 3 GPT-6 Astra max 37.7%
- 4 GPT-6.1 Sol max 35.9%
- 5 Kimi K3.1 Pro max 17%
- 6
DeepSeek V4.1 Flash max 5.66% - 7
Qwen3.8-2.4T-A95B max 1.89%
Source: Official ArXivLean leaderboard
- 1 Claude Opus 5.5 high 47.2%
- 2 Gemini 4 Argon high 39.6%
- 3 GPT-6 Astra max 37.7%
- 4 GPT-6.1 Sol max 35.9%
- 5
Leanstral 1.5 17.1% - 6 Kimi K3.1 Pro max 17%
- 7 DeepSeek V4.1 Flash max 5.66%
- 8 Qwen3.8-2.4T-A95B max 1.89%
Sources: Official ArXivLean leaderboard and labs’ model cards
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Opus 5.5 | 47.2% | high | Sep 2026 | Official leaderboard |
| 2 | Gemini 4 Argon | 39.6% | high | Sep 2026 | Official leaderboard |
| 3 | GPT-6 Astra | 37.7% | max | Sep 2026 | Official leaderboard |
| 4 | GPT-6.1 Sol | 35.9% | max | Sep 2026 | Official leaderboard |
| 5 | Kimi K3.1 Pro | 17% | max | Oct 2026 | Official leaderboard |
| 6 | DeepSeek V4.1 Flash | 5.66% | max | Sep 2026 | Official leaderboard |
| 7 | Qwen3.8-2.4T-A95B | 1.89% | max | Aug 2026 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Opus 5.5 | 47.2% | high | Sep 2026 | Official leaderboard |
| 2 | Gemini 4 Argon | 39.6% | high | Sep 2026 | Official leaderboard |
| 3 | GPT-6 Astra | 37.7% | max | Sep 2026 | Official leaderboard |
| 4 | GPT-6.1 Sol | 35.9% | max | Sep 2026 | Official leaderboard |
| 5 | Leanstral 1.5 | 17.1% | Jul 2026 | Lab model card | |
| 6 | Kimi K3.1 Pro | 17% | max | Oct 2026 | Official leaderboard |
| 7 | DeepSeek V4.1 Flash | 5.66% | max | Sep 2026 | Official leaderboard |
| 8 | Qwen3.8-2.4T-A95B | 1.89% | max | Aug 2026 | Official leaderboard |
Labs report scores for 1 model in their cards. Switch the scores above to rank them together.
3 models have held number one on the official leaderboard since its release in April 2026 3 models have held number one with claims included since its release in April 2026
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official ArXivLean leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Sources: Official ArXivLean leaderboard and labs’ model cards
Labs report it only in system cards
We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark. Some system cards do, below.
One system card reports it
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| Jul 2, 2026 | Leanstral 1.5 technical report | 17.1 | ArXivLean, 7 of 41 problems (3.1.4) | Capabilities |
More in Math
- AIME 2026 In use Models solve the 30 integer-answer problems from the 2026 AIME I and II math competitions. Scored by exact answer match. GPT-5.5 100%
- IMO-AnswerBench In use 400 olympiad problems with short verifiable answers, robustified against memorization. Scored by answer match. GLM-5.2 91%
- HMMT 2026 In use Models solve the 33 final-answer problems from the February 2026 Harvard-MIT Mathematics Tournament. Scored by exact answer match. GPT-5.5 98.5%
- PolyMath In use Models solve 9,000 parallel math problems in 18 languages across four difficulty levels, from school math to frontier math. Final answers are checked. Qwen3.7-Max 86.5%
- MathArena (Apex, IMO, USAMO) In use Evaluates models on newly released math competitions after release. Final-answer or human/LLM-judged proof scores. Claude Opus 4.8 81.3%
- MathArena Apex Shortlist In use Models solve 47 hard final-answer problems that MathArena selected from 2025 competitions, where the best models scored about 50% at selection. Scored by accuracy over repeated attempts. GPT-5.5 98.4%