Frontiermath tiers 1 3 v2 — Epoch AI
Frontiermath tiers 1 3 v2 — Epoch AI is run by Epoch AI. It has ranked 122 models, of which the catalogue holds 85, scoring from 0 to 93.684.
measured by Epoch AI · the board itself
| Place | Model | Metric | Score |
|---|---|---|---|
| 1stof 122 | GPT 6.1 Sol | mean_score (max) | 93.684 |
| 2ndof 122 | GPT 6 Astra | mean_score (max) | 93.684 |
| 3rdof 122 | Claude Opus 5.5 | mean_score (max) | 91.228 |
| 4thof 122 | GPT-6.1 Terra | mean_score (max) | 90.877 |
| 5thof 122 | Claude Fable 5.1 | mean_score (max) | 90.175 |
| 6thof 122 | GPT 6 Sol | mean_score (max) | 89.825 |
| 7thof 122 | GPT-5.6 Sol | mean_score (max) | 89.123 |
| 8thof 122 | Claude Sonnet 5.5 | mean_score (max) | 88.772 |
| 9thof 122 | GPT-5.5 Pro | mean_score (xhigh) | 87.719 |
| 10thof 122 | Claude Fable 5 | mean_score (max) | 87.018 |
| 11thof 122 | Gemini 4 Argon | mean_score (high) | 85.965 |
| 12thof 122 | GPT-5.6 Terra | mean_score (max) | 85.965 |
| 13thof 122 | Claude Opus 5 | mean_score (max) | 85.614 |
| 14thof 122 | GPT-5.5 | mean_score (xhigh) | 85.263 |
| 15thof 122 | GPT-5.4 Pro | mean_score (xhigh) | 82.456 |
| 17thof 122 | GPT-5.6 Luna | mean_score (max) | 82.105 |
| 18thof 122 | Claude Opus 4.8 | mean_score (max) | 80 |
| 19thof 122 | GPT 6 Luna | mean_score (max) | 78.947 |
| 20thof 122 | GPT-5.4 | mean_score (xhigh) | 78.596 |
| 21stof 122 | Claude Haiku 5.5 | mean_score (max) | 75.088 |
| 22ndof 122 | Qwen 3.8 Max | mean_score (xhigh) | 74.737 |
| 23rdof 122 | Muse Spark 1.3 | mean_score (xhigh) | 74.386 |
| 25thof 122 | GPT-5.2 Pro | mean_score (xhigh) | 74 |
| 26thof 122 | Kimi K3 | mean_score (max) | 72.183 |
| 27thof 122 | Gemini 3.7 Flash | mean_score (high) | 71.579 |
| 28thof 122 | DeepSeek V4.1 Flash 1017 | mean_score (max) | 71.228 |
| 29thof 122 | Claude Opus 4.7 | mean_score (max) | 70.175 |
| 30thof 122 | Muse-Spark-Open-816B-A42B | mean_score (xhigh) | 69.825 |
| 31stof 122 | GLM-5.3 | mean_score (max) | 68.772 |
| 32ndof 122 | Gemini 3.8 Flash | mean_score (high) | 68.421 |
| 33rdof 122 | GPT-5.2 | mean_score (xhigh) | 67.4 |
| 34thof 122 | DeepSeek V4.1 Flash | mean_score (max) | 67.368 |
| 35thof 122 | Grok 4.6 | mean_score (xhigh) | 65.965 |
| 36thof 122 | Claude Opus 4.6 | mean_score (max) | 65.965 |
| 38thof 122 | Claude Sonnet 5 | mean_score (max) | 65.614 |
| 39thof 122 | DeepSeek V4 Pro 0813 | mean_score (max) | 64.561 |
| 40thof 122 | Qwen3.7 Max | mean_score (max) | 64.561 |
| 41stof 122 | Gemini 3.5 Flash | mean_score (high) | 62.807 |
| 42ndof 122 | Gemini 3.1 Pro Preview | mean_score | 59.649 |
| 43rdof 122 | GLM 5.2 | mean_score (max) | 59.206 |
| 44thof 122 | Gemini 3.6 Flash | mean_score (high) | 58.947 |
| 46thof 122 | DeepSeek V4 Flash (0731) | mean_score (max) | 57.544 |
| 47thof 122 | Grok 4.5 | mean_score (high) | 57.193 |
| 48thof 122 | Kimi K2.6 | mean_score | 57.193 |
| 49thof 122 | GLM 5.3 Flash | mean_score (max) | 55.789 |
| 50thof 122 | GPT-5 Pro | mean_score (high) | 55.789 |
| 51stof 122 | GPT-5 | mean_score (high) | 55.439 |
| 53rdof 122 | Kimi K2.7 Code | mean_score | 54.035 |
| 54thof 122 | Grok 4.7 | mean_score (xhigh) | 52.982 |
| 55thof 122 | GPT-5.4 mini | mean_score (xhigh) | 51.228 |
| 56thof 122 | Gemini 3 Flash Preview | mean_score | 51.228 |
| 57thof 122 | GPT-5 mini | mean_score (high) | 46.667 |
| 58thof 122 | Inkling Small | mean_score (xhigh) | 46.316 |
| 59thof 122 | DeepSeek V4 Pro | mean_score (max) | 45.263 |
| 60thof 122 | Grok 4.20 | mean_score | 44.912 |
| 61stof 122 | GPT-5.4 nano | mean_score (high) | 44.912 |
| 62ndof 122 | Grok 4.3 | mean_score (high) | 42.807 |
| 66thof 122 | Qwen3.6 Plus | mean_score | 38.246 |
| 68thof 122 | GLM 5.1 | mean_score | 36.842 |
| 69thof 122 | o4-mini | mean_score (high) | 36.14 |
| 70thof 122 | Qwen3.6 27B | mean_score | 35.088 |
| 71stof 122 | Qwen 3.7 Plus | mean_score (none) | 34.386 |
| 72ndof 122 | Claude Opus 4.5 | mean_score | 34.386 |
| 74thof 122 | o3 | mean_score (high) | 33.333 |
| 75thof 122 | Inkling | mean_score (xhigh) | 33.333 |
| 77thof 122 | Qwen3.5 397B A17B | mean_score (none) | 31.228 |
| 81stof 122 | Gemini 3.1 Flash-Lite | mean_score (high) | 27.719 |
| 82ndof 122 | GPT-5.5 Instant | mean_score | 26.316 |
| 83rdof 122 | Gemini 3.5 Flash-Lite | mean_score (high) | 25.965 |
| 86thof 122 | Gemini 2.5 Pro | mean_score | 24.561 |
| 87thof 122 | Claude Sonnet 4.5 | mean_score | 23.86 |
| 89thof 122 | Qwen3.6-Flash | mean_score (none) | 22.456 |
| 91stof 122 | Qwen3.6 35B A3B | mean_score (none) | 20.351 |
| 93rdof 122 | GPT-5 nano | mean_score (high) | 20 |
| 94thof 122 | Qwen3.7 Flash | mean_score (none) | 19.298 |
| 98thof 122 | o3-mini | mean_score (high) | 18.596 |
| 99thof 122 | Qwen3.5-Flash | mean_score (none) | 18.246 |
| 106thof 122 | o1 | mean_score (high) | 14.737 |
| 107thof 122 | Claude Opus 4.1 | mean_score | 12.632 |
| 112thof 122 | GPT-4.1 mini | mean_score | 6.667 |
| 113thof 122 | GPT-4.1 fine-tuned | mean_score | 5.965 |
| 119thof 122 | GPT-4o-mini (2024-07-18) | mean_score | 70.2 |
| 120thof 122 | GPT-4 Turbo | mean_score | 70.2 |
| 121stof 122 | GPT-4o (2024-08-06) | mean_score | 35.1 |
| 122ndof 122 | GPT-3.5 Turbo | mean_score | 0 |
Read from the board on 2026-11-01