DiscoverPhysics - Pass@1: leaderboard
Metric: Expected share of the 22 worlds (%) passed in one attempt, estimated from five seeds per world: a world passes when the submitted law predicts held-out test particles with a variance-normalized trajectory MSE below 0.1 and the explanation scores at least 0.9 against the expert rubric (claude-opus-4-6 judge); higher is better. Source: arxiv.org. Saturation forecast: Around May 2027. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.7 (High) | 26.4 |
| 2 | GPT-5.5 (Medium) | 21.7 |
| 3 | Qwen 3.5 397B A17B | 4.5 |
| 4 | Kimi K2.5 | 2.6 |
| 5 | Claude Sonnet 4.6 | 1.8 |
| 6 | Qwen 3 235B A22B | 1 |
| 7 | GPT-5.4 | 0.9 |
| 8 | Llama 3.3 70B Instruct | 0 |
| 9 | GPT-OSS-120B | 0 |
| 10 | Claude Haiku 4.5 | 0 |
| 11 | GPT-OSS-20B | 0 |
Interactive version: theaggregate.ai/benchmark?slug=discoverphysics-pass-1 · How It Works · Data refreshed daily, snapshot 2026-10-07.