HyperAI
Command Palette
Search for a command to run...
CritPt
A benchmark designed to test LLMs on research-level physics reasoning tasks, featuring 71 composite research challenges.
- Models
- 265 current, 600 in total
- Median
- 0.3%
- Range
- 0.0%–35.7%
| Rank | Compare | Model | Score |
|---|---|---|---|
| 251 |  Qwen3.5 4B (Non-reasoning) Alibaba | 0.0% | |
| 252 | Qwen3.5 Omni Flash Alibaba | 0.0% | |
| 253 | 0.0% | ||
| 254 | Qwen3.8 27B (Low) Alibaba | 0.0% | |
| 255 | Qwen3.8 27B (Medium) Alibaba | 0.0% | |
| 256 | Qwen3 Coder Next Alibaba | 0.0% | |
| 257 | Qwen3 Next 80B A3B Instruct Alibaba | 0.0% | |
| 258 | 0.0% | ||
| 259 | Qwen3 Omni 30B A3B Instruct Alibaba | 0.0% | |
| 260 | 0.0% | ||
| 261 |  Reka Flash 3 Reka AI | 0.0% | |
| 262 |  Sarvam 105B (High) Sarvam | 0.0% | |
| 263 |  Solar Open 100B (Reasoning) Upstage | 0.0% | |
| 264 |  Step3 VL 10B StepFun | 0.0% | |
| 265 |  Tiny Aya Global Cohere | 0.0% |