Critpt — Epoch AI
Critpt — Epoch AI is run by Epoch AI. It has ranked 260 models, of which the catalogue holds 143, scoring from 0 to 85.7.
measured by Epoch AI · the board itself
| Place | Model | Metric | Score |
|---|---|---|---|
| 2ndof 260 | GPT-5.6 Sol | Accuracy (max) | 32.286 |
| 3rdof 260 | GPT 6 Astra | Accuracy (max) | 31.714 |
| 4thof 260 | Claude Opus 5.5 | Accuracy (max) | 31.714 |
| 6thof 260 | GPT 6.1 Sol | Accuracy (max) | 31.714 |
| 9thof 260 | Claude Sonnet 5.5 | Accuracy (max) | 31.429 |
| 10thof 260 | Claude Fable 5.1 | Accuracy (xhigh) | 31.143 |
| 12thof 260 | GPT 6 Sol | Accuracy (max) | 30.857 |
| 14thof 260 | GPT-5.5 Pro | Accuracy (xhigh) | 30.571 |
| 16thof 260 | GPT-5.4 Pro | Accuracy (xhigh) | 30 |
| 17thof 260 | GPT-5.6 Terra | Accuracy (max) | 30 |
| 20thof 260 | Claude Opus 5 | Accuracy (max) | 29.143 |
| 25thof 260 | Claude Fable 5 | Accuracy (max) | 28.571 |
| 29thof 260 | GPT-6.1 Terra | Accuracy (max) | 28 |
| 34thof 260 | GPT-5.5 | Accuracy (xhigh) | 27.143 |
| 36thof 260 | Gemini 4 Argon | Accuracy (high) | 27.143 |
| 38thof 260 | MiMo V2.6 Pro | Accuracy | 26.571 |
| 41stof 260 | Muse Spark 1.3 | Accuracy (xhigh) | 26 |
| 42ndof 260 | Gemini 3 Deep Think | Accuracy | 25.714 |
| 51stof 260 | Kimi K3.1 Pro | Accuracy | 24.571 |
| 52ndof 260 | DeepSeek V4.1 Flash 1017 | Accuracy (max) | 24 |
| 53rdof 260 | GPT-5.4 | Accuracy (xhigh) | 23.429 |
| 54thof 260 | Kimi K3 | Accuracy (max) | 23.4 |
| 58thof 260 | Claude Haiku 5.5 | Accuracy (xhigh) | 22.571 |
| 60thof 260 | Claude Opus 4.8 | Accuracy (max) | 20.857 |
| 61stof 260 | GLM 5.2 | Accuracy (max) | 20.857 |
| 62ndof 260 | Step 5 | Accuracy | 20.857 |
| 63rdof 260 | GPT-5.6 Luna | Accuracy (max) | 20.571 |
| 65thof 260 | Qwen 3.8 Max | Accuracy | 20 |
| 66thof 260 | Grok 4.6 | Accuracy (xhigh) | 19.714 |
| 68thof 260 | GPT 6 Luna | Accuracy (max) | 19.429 |
| 69thof 260 | GLM-5.3 | Accuracy (max) | 19.143 |
| 73rdof 260 | Gemini 3.8 Flash | Accuracy (high) | 18.286 |
| 74thof 260 | DeepSeek V4 Pro 0813 | Accuracy (max) | 18 |
| 75thof 260 | Grok 4.7 | Accuracy (high) | 18 |
| 76thof 260 | Gemini 3.1 Pro Preview | Accuracy | 17.714 |
| 78thof 260 | Muse Spark 1.2 | Accuracy (xhigh) | 17.714 |
| 85thof 260 | Claude Sonnet 5 | Accuracy (max) | 16.857 |
| 87thof 260 | GPT-5.3 Codex | Accuracy (xhigh) | 16.857 |
| 89thof 260 | DeepSeek V4 Flash (0731) | Accuracy (max) | 16.571 |
| 91stof 260 | Grok 4.5 | Accuracy (high) | 15.429 |
| 92ndof 260 | GLM 5.3 Flash | Accuracy | 15.429 |
| 95thof 260 | Muse Spark 1.1 | Accuracy (xhigh) | 15.143 |
| 99thof 260 | Muse-Spark-Open-816B-A42B | Accuracy (xhigh) | 14.857 |
| 101stof 260 | Gemini 3.7 Flash | Accuracy (high) | 14.286 |
| 102ndof 260 | DeepSeek V4.1 Flash | Accuracy (max) | 14.286 |
| 103rdof 260 | Qwen3.7 Max | Accuracy (max) | 13.429 |
| 105thof 260 | Gemini 3.5 Flash | Accuracy (high) | 13.143 |
| 106thof 260 | DeepSeek V4 Pro | Accuracy (max) | 12.857 |
| 108thof 260 | GPT-5 | Accuracy (high) | 12.6 |
| 110thof 260 | Claude Opus 4.7 | Accuracy (max) | 12 |
| 111thof 260 | MiMo V2.6 Flash | Accuracy | 12 |
| 114thof 260 | Gemini 3.6 Flash | Accuracy (high) | 10.571 |
| 117thof 260 | GPT-5.4 mini | Accuracy (xhigh) | 10 |
| 118thof 260 | Kimi K2.7 Code | Accuracy | 10 |
| 122ndof 260 | GPT-5.4 nano | Accuracy (xhigh) | 9.253 |
| 123rdof 260 | Qwen 3.7 Plus | Accuracy | 9.143 |
| 124thof 260 | Grok Build 0.1 | Accuracy | 9.143 |
| 127thof 260 | Inkling Small | Accuracy | 8.286 |
| 129thof 260 | Kimi K2.6 | Accuracy | 8 |
| 130thof 260 | Grok 4.3 | Accuracy (high) | 8 |
| 131stof 260 | DeepSeek V4 Flash | Accuracy (max) | 7.143 |
| 132ndof 260 | gemini-3-pro-preview | Accuracy | 6.9 |
| 135thof 260 | Inkling | Accuracy (xhigh) | 5.429 |
| 136thof 260 | Qwen3.8 27B | Accuracy (xhigh) | 5.429 |
| 138thof 260 | GPT-5.1 | Accuracy | 4.857 |
| 141stof 260 | GLM 5.1 | Accuracy | 4.571 |
| 143rdof 260 | MiMo-V2.5-Pro | Accuracy | 4 |
| 146thof 260 | MiniMax M3 | Accuracy | 3.714 |
| 147thof 260 | Ring-2.6-1T | Accuracy | 3.714 |
| 148thof 260 | MiMo-V2.5 | Accuracy | 3.714 |
| 150thof 260 | Nemotron 3 Ultra | Accuracy | 3.143 |
| 151stof 260 | Claude Sonnet 4.6 | Accuracy (max) | 3.143 |
| 152ndof 260 | Nemotron 3 Super | Accuracy | 3.143 |
| 153rdof 260 | Kimi K2.5 | Accuracy | 3.143 |
| 156thof 260 | Qwen3.6 Plus | Accuracy | 2.857 |
| 161stof 260 | Step 3.7 Flash | Accuracy | 2.286 |
| 162ndof 260 | Gemini 2.5 Pro | Accuracy | 2 |
| 164thof 260 | DeepSeek V3.1 Terminus | Accuracy | 1.714 |
| 165thof 260 | GLM-4.7 | Accuracy | 1.71 |
| 166thof 260 | GPT OSS 20B | Accuracy (high) | 1.429 |
| 168thof 260 | Gemma 4 31B | Accuracy | 1.429 |
| 169thof 260 | o3 | Accuracy (high) | 1.4 |
| 170thof 260 | Gemini 3.1 Flash-Lite | Accuracy | 1.143 |
| 171stof 260 | GPT OSS 120B | Accuracy (high) | 1.143 |
| 172ndof 260 | Claude Sonnet 4.5 | Accuracy | 1.143 |
| 173rdof 260 | GLM-4.6 | Accuracy | 1.143 |
| 176thof 260 | DeepSeek-R1 | Accuracy | 1.1 |
| 177thof 260 | Gemini 2.5 Flash | Accuracy | 1.1 |
| 178thof 260 | Trinity Large Thinking | Accuracy | 85.7 |
| 179thof 260 | Qwen3.6 27B | Accuracy (none) | 85.7 |
| 180thof 260 | Qwen3.5-122B-A10B | Accuracy (none) | 85.7 |
| 181stof 260 | Qwen3.5 397B A17B | Accuracy (none) | 85.7 |
| 182ndof 260 | Mercury 2 | Accuracy | 84.9 |
| 183rdof 260 | o4-mini | Accuracy (high) | 60 |
| 184thof 260 | MiniMax M2.7 | Accuracy | 57.1 |
| 185thof 260 | Qwen3.5-35B-A3B | Accuracy (none) | 57.1 |
| 186thof 260 | Qwen3.5 9B | Accuracy (none) | 57.1 |
| 187thof 260 | Claude Opus 4 | Accuracy | 30 |
| 189thof 260 | Qwen3.6 35B A3B | Accuracy | 28.6 |
| 190thof 260 | Claude Sonnet 4 | Accuracy | 28.6 |
| 191stof 260 | Qwen3 32B | Accuracy | 28.6 |
| 194thof 260 | Command A Plus | Accuracy | 28.6 |
| 195thof 260 | o3-mini | Accuracy (high) | 28.6 |
| 198thof 260 | Qwen3 30B A3B Thinking 2507 | Accuracy | 28.6 |
| 200thof 260 | Llama-4-Maverick-17B-128E-Instruct | Accuracy | 0 |
| 202ndof 260 | GPT-4o (2024-11-20) | Accuracy | 0 |
| 205thof 260 | Claude Haiku 4.5 | Accuracy | 0 |
| 206thof 260 | GPT-4.1 nano | Accuracy | 0 |
| 207thof 260 | MiMo-V2-Flash | Accuracy | 0 |
| 208thof 260 | Gemma 4 26B A4B | Accuracy | 0 |
| 209thof 260 | GPT-4.1 mini | Accuracy | 0 |
| 210thof 260 | Gemma 3 27B | Accuracy | 0 |
| 211thof 260 | Llama 3.1 8B Instruct | Accuracy | 0 |
| 212thof 260 | Claude Haiku 3.5 | Accuracy | 0 |
| 213thof 260 | mistral-small-2503 | Accuracy | 0 |
| 214thof 260 | DeepSeek-V3 | Accuracy | 0 |
| 216thof 260 | DeepSeek-V3 0324 | Accuracy | 0 |
| 217thof 260 | Gemini 3.5 Flash-Lite | Accuracy | 0 |
| 218thof 260 | Llama 3.3 70B Instruct | Accuracy | 0 |
| 222ndof 260 | Mistral Large 3 | Accuracy | 0 |
| 224thof 260 | Solar Pro 3 | Accuracy | 0 |
| 225thof 260 | Phi-4-mini-instruct | Accuracy | 0 |
| 226thof 260 | Qwen3 Coder Next | Accuracy | 0 |
| 227thof 260 | granite-4.1-30b | Accuracy | 0 |
| 229thof 260 | Gemma 3 12B | Accuracy | 0 |
| 230thof 260 | Qwen3 235B A22B Thinking 2507 | Accuracy | 0 |
| 231stof 260 | GPT-5 mini | Accuracy | 0 |
| 234thof 260 | GPT-5.5 Instant | Accuracy | 0 |
| 240thof 260 | Qwen3 8B | Accuracy | 0 |
| 241stof 260 | Qwen3 14B | Accuracy | 0 |
| 244thof 260 | Mercury 2.5 | Accuracy | 0 |
| 246thof 260 | Qwen3.5-4B | Accuracy (none) | 0 |
| 247thof 260 | gemma-3-270m-it | Accuracy | 0 |
| 248thof 260 | Nova Micro | Accuracy | 0 |
| 250thof 260 | Granite 4.0 Micro | Accuracy | 0 |
| 251stof 260 | Qwen3.5-2B | Accuracy (none) | 0 |
| 253rdof 260 | c4ai-command-a-03-2025 | Accuracy | 0 |
| 254thof 260 | Qwen3.5-0.8B | Accuracy | 0 |
| 255thof 260 | Llama 3.2 11B Vision Instruct | Accuracy | 0 |
| 256thof 260 | Meta-Llama-3.1-405B-Instruct | Accuracy | 0 |
| 258thof 260 | Nemotron Nano 9B V2 | Accuracy | 0 |
| 259thof 260 | Reka Flash 3 | Accuracy | 0 |
| 260thof 260 | Phi-4 | Accuracy | 0 |
Read from the board on 2026-10-31