Reasoning
CritPt Leaderboard
CritPt (Critical Points) tests graduate-level physics problem-solving on research-style questions spanning mechanics, electromagnetism, quantum mechanics, and more. It was created by physicists to probe whether a model can reason through multi-step physics rather than pattern-match textbook answers.
Source: epoch50 open models ranked+117 proprietaryData through Sep 2026
Open models ranked on CritPt
# shows rank among open models / rank overall (including proprietary).
| # | Model | Score |
|---|---|---|
| 1 / 27 | Kimi K3 · 2779.9B | 23.4% |
| 2 / 32 | GLM 5.2 · 753.3B | 20.9% |
| 3 / 38 | GLM 5.3 · 753.3B | 19.1% |
| 4 / 41 | DeepSeek V4 Pro 0813 · 1650.5B | 18.0% |
| 5 / 48 | DeepSeek V4 Flash 0731 · 304.2B | 16.6% |
| 6 / 50 | GLM 5.3 Flash · 321.3B | 15.4% |
| 7 / 57 | DeepSeek V4 Pro · 1598.8B | 12.9% |
| 8 / 64 | Kimi K2.7 Code · 1026.9B | 10.0% |
| 9 / 70 | Inkling Small · 266.0B | 8.3% |
| 10 / 73 | Kimi K2.6 · 1026.9B | 8.0% |
| 11 / 74 | DeepSeek V4 Flash · 290.9B | 7.1% |
| 12 / 78 | Inkling · 952.4B | 5.4% |
| 13 / 79 | Qwen3.8 27B · 27.8B | 5.4% |
| 14 / 83 | GLM 5.1 · 753.9B | 4.6% |
| 15 / 85 | MiMo V2.5 Pro · 1023.2B | 4.0% |
| 16 / 86 | MiMo V2.5 · 310.8B | 3.7% |
| 17 / 87 | MiniMax M3 · 427.0B | 3.7% |
| 18 / 88 | Ring 2.6 1T · 1025.7B | 3.7% |
| 19 / 90 | Kimi K2.5 · 1026.9B | 3.1% |
| 20 / 93 | DeepSeek V3.2 Exp · 685.4B | 2.9% |
| 21 / 96 | Step 3.7 Flash · 201.4B | 2.3% |
| 22 / 100 | GLM 4.7 · 358.3B | 1.7% |
| 23 / 101 | Gemma 4 31B IT · 31.3B | 1.4% |
| 24 / 103 | GPT OSS 20B · 20.9B | 1.4% |
| 25 / 108 | GLM 4.6 · 356.8B | 1.1% |
| 26 / 109 | GPT OSS 120B · 116.8B | 1.1% |
| 27 / 110 | DeepSeek R1 · 684.5B | 1.1% |
| 28 / 112 | Qwen3.5 122B A10B · 125.1B | 0.9% |
| 29 / 113 | Qwen3.6 27B · 27.8B | 0.9% |
| 30 / 114 | Trinity Large Thinking · 398.6B | 0.9% |
| 31 / 117 | MiniMax M2.7 · 228.7B | 0.6% |
| 32 / 126 | Qwen3 30B A3B Thinking 2507 · 30.5B | 0.3% |
| 33 / 127 | Qwen3 32B · 32.8B | 0.3% |
| 34 / 128 | Qwen3.5 9B · 9.7B | 0.3% |
| 35 / 129 | Qwen3.6 35B A3B · 36.0B | 0.3% |
| 36 / 133 | DeepSeek v3 · 684.5B | 0.0% |
| 37 / 134 | DeepSeek v3 0324 · 684.5B | 0.0% |
| 38 / 137 | Gemma 3 12B IT · 12.2B | 0.0% |
| 39 / 138 | Gemma 3 27B IT · 27.4B | 0.0% |
| 40 / 146 | Granite 4.1 30B · 28.9B | 0.0% |
| 41 / 149 | Llama 3.1 8B Instruct · 8.0B | 0.0% |
| 42 / 150 | Llama 3.3 70B Instruct · 70.6B | 0.0% |
| 43 / 151 | Llama 4 Maverick 17B 128E Instruct · 401.6B | 0.0% |
| 44 / 152 | Llama 4 Scout 17B 16E Instruct · 108.6B | 0.0% |
| 45 / 153 | MiMo v2 Flash · 309.8B | 0.0% |
| 46 / 162 | Phi 4 Mini Instruct · 3.8B | 0.0% |
| 47 / 163 | Qwen3 14B · 14.8B | 0.0% |
| 48 / 164 | Qwen3 235B A22B Thinking 2507 · 235.1B | 0.0% |
| 49 / 165 | Qwen3 8B · 8.2B | 0.0% |
| 50 / 166 | Qwen3 Coder Next · 79.7B | 0.0% |
Score vs model size
Which models give the most quality for their size — the ones worth running locally.
- Phi 4 Mini Instruct, 4B, score 0.0% — on the efficiency frontier (best score at its size or smaller).
- Qwen3.5 9B, 10B, score 0.3% — on the efficiency frontier (best score at its size or smaller).
- GPT OSS 20B, 21B, score 1.4% — on the efficiency frontier (best score at its size or smaller).
- Qwen3.8 27B, 28B, score 5.4% — on the efficiency frontier (best score at its size or smaller).
- Inkling Small, 266B, score 8.3% — on the efficiency frontier (best score at its size or smaller).
- DeepSeek V4 Flash 0731, 304B, score 16.6% — on the efficiency frontier (best score at its size or smaller).
- GLM 5.2, 753B, score 20.9% — on the efficiency frontier (best score at its size or smaller).
- Kimi K3, 2.8T, score 23.4% — on the efficiency frontier (best score at its size or smaller).
CritPt: frequently asked questions
- What is the best open LLM on CritPt?
- Kimi K3 is the top open model on CritPt, scoring 23.4%. Among all models tested — including proprietary ones — it ranks #27. The top model overall is GPT 5.6 Sol Max (OpenAI) at 32.3%.
- What's the best CritPt model you can run on a 24 GB GPU?
- Qwen3.8 27B is the highest-scoring open model that fits in 24 GB at 4-bit quantization (about 15 GB), scoring 5.4% on CritPt.
- What's the best CritPt model you can run on a 12 GB GPU?
- GPT OSS 20B is the highest-scoring open model that fits in 12 GB at 4-bit quantization (about 12 GB), scoring 1.4% on CritPt.
- Can open models match proprietary models on CritPt?
- Not quite on CritPt: the strongest proprietary model (GPT 5.6 Sol Max) scores 32.3%, ahead of the best open model (Kimi K3) at 23.4% — but you can run the open one yourself.
Scores aggregated from epoch. llmrun does not run this benchmark — see the source for methodology, or the about benchmarks for what it measures.