Benchmarks
Enigma Eval
Models scored
42
Best score
0.24
Metric
Accuracy
Enigma Eval is an AI evaluation tracked by Epoch AI, with 42 scored model versions on record. The scores here are reported by third parties rather than produced by Epoch AI running the evaluation itself.
Scored models span February 2024 to March 2026. The best score recorded is 0.24, measured as Accuracy. Models come from OpenAI, Google DeepMind, Anthropic, Moonshot, Google, Google DeepMind,Google, Mistral AI, Meta AI.
Full record
- Score metric
- Accuracy
- Models scored
- 42
- Best score recorded
- 0.24
- Earliest model scored
- Feb 29, 2024
- Most recent model scored
- Mar 5, 2026
- Third-party reported
- Yes
Leaderboard
| 01 | gpt-5.4-pro-2026-03-05_unknown | OpenAI | 0.24 |
| 02 | gemini-3.1-pro-preview | Google DeepMind | 0.2 |
| 03 | gpt-5-pro-2025-10-06_unknown | OpenAI | 0.19 |
| 04 | gemini-3-pro-preview | Google DeepMind | 0.18 |
| 05 | gpt-5.4-2026-03-05_xhigh | OpenAI | 0.16 |
| 06 | o3-2025-04-16_medium | OpenAI | 0.13 |
| 07 | o3-2025-04-16_high | OpenAI | 0.12 |
| 08 | claude-opus-4-5-20251101_unknown | Anthropic | 0.12 |
| 09 | gpt-5.1-2025-11-13_unknown | OpenAI | 0.11 |
| 10 | gpt-5-2025-08-07_unknown | OpenAI | 0.1 |
| 11 | gpt-5.2-2025-12-11_unknown | OpenAI | 0.1 |
| 12 | o4-mini-2025-04-16_high | OpenAI | 0.09 |
| 13 | gpt-5-mini-2025-08-07_unknown | OpenAI | 0.08 |
| 14 | claude-opus-4-6_max | Anthropic | 0.08 |
| 15 | claude-opus-4-1-20250805_unknown | Anthropic | 0.07 |
| 16 | claude-opus-4-6 | Anthropic | 0.07 |
| 17 | o4-mini-2025-04-16_medium | OpenAI | 0.07 |
| 18 | o1-pro-2025-03-19 | OpenAI | 0.06 |
| 19 | claude-sonnet-4-5-20250929_unknown | Anthropic | 0.06 |
| 20 | o1-2024-12-17_unknown | OpenAI | 0.06 |
| 21 | gemini-2.5-pro-preview-06-05 | Google DeepMind | 0.06 |
| 22 | claude-opus-4-20250514_unknown | Anthropic | 0.06 |
| 23 | claude-3-7-sonnet-20250219_unknown | Anthropic | 0.04 |
| 24 | gemini-2.5-pro-exp-03-25 | Google DeepMind | 0.04 |
| 25 | kimi-k2.5 | Moonshot | 0.03 |
| 26 | gpt-4.5-preview-2025-02-27 | OpenAI | 0.03 |
| 27 | claude-sonnet-4-20250514_unknown | Anthropic | 0.03 |
| 28 | gemini-3.1-flash-lite | 0.03 | |
| 29 | gemini-2.5-flash-preview-05-20 | Google DeepMind | 0.03 |
| 30 | gemini-2.5-pro-preview-05-06 | Google DeepMind | 0.02 |
| 31 | claude-3-7-sonnet-20250219 | Anthropic | 0.02 |
| 32 | gpt-4.1-2025-04-14 | OpenAI | 0.02 |
| 33 | gpt-5.1-2025-11-13_none | OpenAI | 0.02 |
| 34 | gemini-2.0-flash-thinking-exp-01-21 | Google DeepMind,Google | 0.01 |
| 35 | claude-3-5-sonnet-20241022 | Anthropic | 0.01 |
| 36 | pixtral-large-2411 | Mistral AI | 0.01 |
| 37 | claude-3-opus-20240229 | Anthropic | 0.01 |
| 38 | gpt-4o-2024-11-20 | OpenAI | 0.01 |
| 39 | gemini-2.0-pro-exp-02-05 | Google DeepMind | 0.01 |
| 40 | gemini-2.0-flash-02-05 | Google DeepMind,Google | 0.01 |
| 41 | Llama-4-Maverick-17B-128E-Instruct | Meta AI | 0.01 |
| 42 | Llama-3.2-90B-Vision-Instruct | Meta AI | 0 |
More benchmarks
SourceEpoch AI, 'AI Benchmarking Hub'. Published online at epoch.ai. Retrieved 2026-07-29 from https://epoch.ai/benchmarks. Licensed under CC BY 4.0.
← All benchmarks