Skip to content
Models measured
45
Scores recorded
49
Sources
1
IQ weight
1.36

Score distribution

3,629.7
Mean
3,843
Median
580.2
Std dev
2,719
10th percentile
4,229
90th percentile
2,131
Min
2,1314,404
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.36 for chat models, measured over 45 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 data4,404
2Gemini 3 Flash· Epoch AI Benchmarking Hub data4,333
3Gemini 3.8 Flash· Epoch AI Benchmarking Hub data4,307
4Claude Fable 5.1high· Epoch AI Benchmarking Hub data4,303
5Gemini 3.5 Flash· Epoch AI Benchmarking Hub data4,288
6Gemini 3.1 Flash Lite· Epoch AI Benchmarking Hub data4,229
7Gemini 3.1 Pro Preview· Epoch AI Benchmarking Hub data4,172
8Gemini 3 Pro Preview· Epoch AI Benchmarking Hub data4,145
9Claude Opus 4.8high· Epoch AI Benchmarking Hub data4,128
10Muse Spark 1.3· Epoch AI Benchmarking Hub data4,103
11Gemini 2.5 Pro· Epoch AI Benchmarking Hub data4,093
12Gemini 2.5 Pro Preview 05-06· Epoch AI Benchmarking Hub data4,085
13Gemini 2.5 Pro· Epoch AI Benchmarking Hub data4,019
14GPT 5medium· Epoch AI Benchmarking Hub data3,937
15gemini-1.5-flash-002· Epoch AI Benchmarking Hub data3,934
16OpenAI o3high· Epoch AI Benchmarking Hub data3,930
17o1medium· Epoch AI Benchmarking Hub data3,925
18Gemini 2.0 Flash· Epoch AI Benchmarking Hub data3,897
19gemini-2.0-pro-exp-02-05· Epoch AI Benchmarking Hub data3,884
20Claude Sonnet 5high· Epoch AI Benchmarking Hub data3,878
21Claude Opus 4.532k· Epoch AI Benchmarking Hub data3,877
22OpenAI o3medium· Epoch AI Benchmarking Hub data3,858
23gemini-2.5-flash-preview-04-17· Epoch AI Benchmarking Hub data3,854
24Claude Sonnet 3.715k· Epoch AI Benchmarking Hub data3,844
25GPT-5.6 Solhigh· Epoch AI Benchmarking Hub data3,843
26Claude Sonnet 3.7· Epoch AI Benchmarking Hub data3,791
27GPT 4.1· Epoch AI Benchmarking Hub data3,791
28Gemini 2.5 Flash Preview 05-20 (no think)· Epoch AI Benchmarking Hub data3,786
29Gemini 2.0 Flash Experimentalthinking· Epoch AI Benchmarking Hub data3,767
30Claude Sonnet 4.632k· Epoch AI Benchmarking Hub data3,762
31o4 Minihigh· Epoch AI Benchmarking Hub data3,717
32GPT 4o· Epoch AI Benchmarking Hub data3,696
33o4 Minimedium· Epoch AI Benchmarking Hub data3,651
34Grok 4.7· Epoch AI Benchmarking Hub data3,525
35Qwen2.5 VL 72B Instruct· Epoch AI Benchmarking Hub data3,448
36GPT-4o mini· Epoch AI Benchmarking Hub data3,401
37Claude 3.5 Sonnet· Epoch AI Benchmarking Hub data3,327
38Claude Sonnet 4.532k· Epoch AI Benchmarking Hub data3,252
39Grok 4.3· Epoch AI Benchmarking Hub data3,229
40Gemma 3 27B IT· Epoch AI Benchmarking Hub data3,087
41Llama 4 Maverick· Epoch AI Benchmarking Hub data3,069
42Claude Opus 432k· Epoch AI Benchmarking Hub data2,915
43Llama 3.2 90B Vision Instruct· Epoch AI Benchmarking Hub data2,798
44Grok 4· Epoch AI Benchmarking Hub data2,719
45Claude Haiku 3.5· Epoch AI Benchmarking Hub data2,507
46Claude Sonnet 4· Epoch AI Benchmarking Hub data2,468
47Qwen 3.8 27B· Epoch AI Benchmarking Hub data2,377
48Claude Sonnet 432k· Epoch AI Benchmarking Hub data2,369
49Pixtral 12B2409· Epoch AI Benchmarking Hub data2,131

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