PhyX-AF - PhyX Modern Physics (Image): leaderboard
Metric: Human-verified autoformalization accuracy (%; 7 PhyX quantum-mechanics and relativity problems with the diagram as image input, Lean 4 output judged valid by two expert annotators). Source: arxiv.org. Saturation forecast: Around December 2026. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 | 42.9 |
| 2 | Gemini 3 Pro | 14.3 |
| 3 | Gemini 2.5 Pro | 0 |
| 4 | Qwen 3 VL 235B A22B Instruct | 0 |
| 5 | Qwen 2.5 VL 72B Instruct | 0 |
Interactive version: theaggregate.ai/benchmark?slug=phyx-af-phyx-modern-physics-image · How It Works · Data refreshed daily, snapshot 2026-09-25.