Skip to content

Benchmark

WeirdML

Scores are percentages.

Models measured
129
Scores recorded
164
Sources
1
IQ weight
2.56

Score distribution

49.8%
Mean
46.1%
Median
21.7%
Std dev
22.2%
10th percentile
82.9%
90th percentile
1.7%
Min
1.7%93.6%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.56 for chat models, measured over 129 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1GPT-6 Astrapro / max· Epoch AI Benchmarking Hub data93.6%
2GPT-6 Astramax· Epoch AI Benchmarking Hub data93.3%
3Claude Fable 5.1max· Epoch AI Benchmarking Hub data92.9%
4GPT-6 Astrahigh· Epoch AI Benchmarking Hub data92.9%
5Claude Fable 5.1high· Epoch AI Benchmarking Hub data92.3%
6Claude Fable 5max· Epoch AI Benchmarking Hub data91.9%
7Claude Opus 5max· Epoch AI Benchmarking Hub data91.8%
8Claude Opus 5high· Epoch AI Benchmarking Hub data91.6%
9GPT-5.6 Solpro / max· Epoch AI Benchmarking Hub data89.4%
10GPT-5.6 Solhigh· Epoch AI Benchmarking Hub data88.8%
11Claude Fable 5high· Epoch AI Benchmarking Hub data87.9%
12GPT-5.6 Solmax· Epoch AI Benchmarking Hub data87%
13Claude Opus 5· Epoch AI Benchmarking Hub data86.3%
14GPT 5.5xhigh· Epoch AI Benchmarking Hub data84.9%
15Gemini 3.8 Flashhigh· Epoch AI Benchmarking Hub data84.8%
16GPT 5.5high· Epoch AI Benchmarking Hub data83.9%
17Claude Opus 4.8xhigh· Epoch AI Benchmarking Hub data82.9%
18Kimi K3max· Epoch AI Benchmarking Hub data82.6%
19GPT-5.3 Codex· Epoch AI Benchmarking Hub data79.3%
20GPT-5.6 Terrahigh· Epoch AI Benchmarking Hub data78.3%
21Claude Opus 4.6high· Epoch AI Benchmarking Hub data78%
22Claude Opus 4.6· Epoch AI Benchmarking Hub data77.9%
23GPT-5.3 Codexxhigh· Epoch AI Benchmarking Hub data77.9%
24GPT 5.4xhigh· Epoch AI Benchmarking Hub data77.7%
25Claude Opus 4.7high· Epoch AI Benchmarking Hub data76.4%
26Claude Opus 4.7· Epoch AI Benchmarking Hub data76.4%
27Claude Opus 4.8medium· Epoch AI Benchmarking Hub data76%
28Claude Opus 4.7max· Epoch AI Benchmarking Hub data75.5%
29GLM 5.3max· Epoch AI Benchmarking Hub data75.4%
30GPT 5.2xhigh· Epoch AI Benchmarking Hub data72.2%
31Gemini 3.1 Pro Preview· Epoch AI Benchmarking Hub data72.1%
32Claude Opus 4.8none· Epoch AI Benchmarking Hub data70.5%
33GLM 5.2max· Epoch AI Benchmarking Hub data70.1%
34Gemini 3 Pro Preview· Epoch AI Benchmarking Hub data69.9%
35Claude Sonnet 5high· Epoch AI Benchmarking Hub data69%
36GLM 5.2high· Epoch AI Benchmarking Hub data67.3%
37Grok 4.6high· Epoch AI Benchmarking Hub data67.3%
38GPT 5.5none· Epoch AI Benchmarking Hub data67.2%
39DeepSeek V4 Pro 0813max· Epoch AI Benchmarking Hub data66.2%
40Claude Sonnet 4.6medium· Epoch AI Benchmarking Hub data66.1%
41Claude Opus 4.516k· Epoch AI Benchmarking Hub data63.7%
42GPT 5.2medium· Epoch AI Benchmarking Hub data63.4%
43DeepSeek V4 Flash 0731max· Epoch AI Benchmarking Hub data63%
44Gemini 3.5 Flashhigh· Epoch AI Benchmarking Hub data62.6%
45Gemini 3 Flash· Epoch AI Benchmarking Hub data61.6%
46GPT-5.6 Lunahigh· Epoch AI Benchmarking Hub data60.9%
47GPT 5.1high· Epoch AI Benchmarking Hub data60.8%
48GPT 5high· Epoch AI Benchmarking Hub data60.7%
49GPT-5 Prohigh· Epoch AI Benchmarking Hub data60.4%
50Muse Spark 1.2xhigh· Epoch AI Benchmarking Hub data60.4%

Showing 1–50 of 164 results.

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