Epoch AI - Enigma Eval: leaderboard
Epoch AI mirror of Scale AI's EnigmaEval — long multimodal reasoning challenges drawn from puzzle-hunt competitions, with unambiguous verifiable solutions; accuracy (%).
Metric: Score. Source: epoch.ai. 46 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | claude-fable-5-high | 39.28 | |
| 2 | gpt-5.6-sol-high | 37.12 | |
| 3 | gemini-3.1-pro-preview-high | 36.78 | |
| 4 | gemini-3.5-flash-high | 25.41 | |
| 5 | gpt-5.4-pro-2026-03-05 | 23.82 | |
| 6 | claude-opus-4-8-xhigh | 23.51 | |
| 7 | gpt-5-pro-2025-10-06 | 18.75 | |
| 8 | gemini-3-pro-preview | 18.24 | |
| 9 | gpt-5.4-2026-03-05 (xhigh thinking) | 15.96 | |
| 10 | o3 (medium) (April 2025) | 13.09 |
Interactive version: theaggregate.ai/benchmark?slug=epoch-ai-enigma-eval · How It Works · Data refreshed daily, snapshot 2026-10-20.