ACR Human Audit - At Least Three Correct Methods: leaderboard
The ACR study uses human reviewers to assess the correctness and variety of mathematical solutions from two unmodified models, with four candidate solutions per problem. Mean problems with at least three correct primary methods (of30).
Metric: Mean problems with at least three correct primary methods (of30). Source: arxiv.org. Saturation forecast: Around 2029. 2 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemma 4 26B A4B | 1 | #353 |
| 2 | Qwen 3.6 35B A3B (Non-reasoning) | 1 | #233 (Qwen 3.6 35B A3B) |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=acr-human-audit-at-least-three-correct-methods · How It Works · Data refreshed daily, snapshot 2026-10-29.