Skip to content

Benchmark

Enigma Eval

Scores are percentages.

Models measured
41
Scores recorded
45
Sources
1
IQ weight
2.48

Score distribution

9.3%
Mean
6%
Median
10%
Std dev
0.8%
10th percentile
23.8%
90th percentile
0.38%
Min
0.38%39.3%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.48 for chat models, measured over 41 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Claude Fable 5high· Epoch AI Benchmarking Hub data39.3%
2GPT-5.6 Solhigh· Epoch AI Benchmarking Hub data37.1%
3Gemini 3.1 Pro Preview· Epoch AI Benchmarking Hub data36.8%
4Gemini 3.5 Flashhigh· Epoch AI Benchmarking Hub data25.4%
5GPT-5.4 Pro· Epoch AI Benchmarking Hub data23.8%
6Claude Opus 4.8xhigh· Epoch AI Benchmarking Hub data23.5%
7GPT-5 Pro· Epoch AI Benchmarking Hub data18.8%
8Gemini 3 Pro Preview· Epoch AI Benchmarking Hub data18.2%
9GPT 5.4xhigh· Epoch AI Benchmarking Hub data16%
10OpenAI o3medium· Epoch AI Benchmarking Hub data13.1%
11Claude Opus 4.5· Epoch AI Benchmarking Hub data11.9%
12OpenAI o3high· Epoch AI Benchmarking Hub data11.9%
13GPT 5.1· Epoch AI Benchmarking Hub data11.2%
14GPT 5· Epoch AI Benchmarking Hub data10.5%
15GPT 5.2· Epoch AI Benchmarking Hub data10.4%
16o4 Minihigh· Epoch AI Benchmarking Hub data9.2%
17GPT-5 mini· Epoch AI Benchmarking Hub data8.2%
18Claude Opus 4.6max· Epoch AI Benchmarking Hub data7.6%
19Claude Opus 4.1· Epoch AI Benchmarking Hub data7.2%
20Claude Opus 4.6· Epoch AI Benchmarking Hub data6.8%
21o4 Minimedium· Epoch AI Benchmarking Hub data6.8%
22O1 Pro· Epoch AI Benchmarking Hub data6.1%
23Claude Sonnet 4.5· Epoch AI Benchmarking Hub data6%
24o1· Epoch AI Benchmarking Hub data5.7%
25Claude Opus 4· Epoch AI Benchmarking Hub data5.6%
26Gemini 2.5 Pro· Epoch AI Benchmarking Hub data5.6%
27Claude Sonnet 3.7· Epoch AI Benchmarking Hub data4.2%
28Gemini 2.5 Pro· Epoch AI Benchmarking Hub data4.1%
29Kimi K2.5· Epoch AI Benchmarking Hub data3.4%
30GPT-4.5 Preview· Epoch AI Benchmarking Hub data3.2%
31Claude Sonnet 4· Epoch AI Benchmarking Hub data3.1%
32Gemini 3.1 Flash Lite· Epoch AI Benchmarking Hub data3%
33Gemini 2.5 Flash Preview 05-20 (no think)· Epoch AI Benchmarking Hub data2.7%
34Gemini 2.5 Pro Preview 05-06· Epoch AI Benchmarking Hub data2.4%
35GPT 4.1· Epoch AI Benchmarking Hub data2.2%
36GPT 5.1none· Epoch AI Benchmarking Hub data1.9%
37Gemini 2.0 Flash Experimentalthinking· Epoch AI Benchmarking Hub data1.1%
38Claude 3.5 Sonnet· Epoch AI Benchmarking Hub data0.91%
39Pixtral Large2411· Epoch AI Benchmarking Hub data0.84%
40Claude 3 Opus· Epoch AI Benchmarking Hub data0.82%
41GPT 4o· Epoch AI Benchmarking Hub data0.8%
42gemini-2.0-pro-exp-02-05· Epoch AI Benchmarking Hub data0.69%
43Gemini 2.0 Flash· Epoch AI Benchmarking Hub data0.63%
44Llama 4 Maverick· Epoch AI Benchmarking Hub data0.58%
45Llama 3.2 90B Vision Instruct· Epoch AI Benchmarking Hub data0.38%

Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence)