CritPt
Science · 2025-09-30
70 research-level physics challenges evaluated by Artificial Analysis.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-5.5 Pro | 30.6 |
| GPT-5.4 Pro | 30.0 |
| GPT-5.6 Terra | 30.0 |
| Claude Opus 5 | 29.1 |
| Claude Fable 5 | 28.6 |
| GPT-5.5 | 27.1 |
| Gemini 3 Deep Think | 25.7 |
| GPT-5.6 Sol | 25.7 |
| GPT-5.4 | 23.4 |
| Opus 4.8 | 20.9 |
| GLM-5.2 | 20.9 |
| GPT-5.6 Luna | 20.6 |
| Gemini 3.1 Pro Preview | 17.7 |
| Sonnet 5 | 16.9 |
| GPT-5.3-Codex | 16.9 |
| DeepSeek-V4-Flash-0731 | 16.6 |
| Grok 4.5 | 15.4 |
| Muse Spark 1.1 | 15.1 |
| Qwen3.7-Max | 13.4 |
| Gemini 3.5 Flash | 13.1 |
| DeepSeek-V4-Pro | 12.9 |
| Opus 4.6 | 12.6 |
| Opus 4.7 | 12.0 |
| GPT-5.2 | 11.6 |
| Muse Spark | 11.3 |
| Agnes 2.5 Pro Alpha | 10.9 |
| Gemini 3.6 Flash | 10.6 |
| GPT-5.4 Mini | 10.0 |
| Kimi K2.7 Code | 10.0 |
| GPT-5.4 Nano | 9.3 |
| Gemini 3 Pro Preview | 9.1 |
| Grok Build 0.1 | 9.1 |
| Qwen3.7-Plus | 9.1 |
| GPT-5.2-Codex | 8.7 |
| Gemini 3 Flash Preview | 8.6 |
| Nex-N2-Pro | 8.6 |
| Grok 4.3 | 8.0 |
| Kimi K2.6 | 8.0 |
| DeepSeek-V3.2-Speciale | 7.4 |
| DeepSeek-V4-Flash | 7.1 |