Skip to content

Benchmark

SimpleBench

Scores are percentages.

Models measured
105
Scores recorded
114
Sources
1
IQ weight
1.75

Score distribution

54.6%
Mean
56.7%
Median
20.9%
Std dev
23.5%
10th percentile
81.9%
90th percentile
10.7%
Min
10.7%94.9%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.75 for chat models, measured over 105 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 5.2· Epoch AI Benchmarking Hub data94.9%
2Gemini 4 Neon· Epoch AI Benchmarking Hub data94.4%
3Gemini 4 Carbon· Epoch AI Benchmarking Hub data92.6%
4Gemini 4 Argon· Epoch AI Benchmarking Hub data91.7%
5Gemini 4 Hydrogen· Epoch AI Benchmarking Hub data91.7%
6Claude Opus 5.5· Epoch AI Benchmarking Hub data88.4%
7Claude Fable 5.1· Epoch AI Benchmarking Hub data86.6%
8GPT-6 Astrapro / max· Epoch AI Benchmarking Hub data86.5%
9GPT-6 Astra· Epoch AI Benchmarking Hub data83.6%
10GPT-6.1 Sol· Epoch AI Benchmarking Hub data82.9%
11Gemini 3.8 Flash· Epoch AI Benchmarking Hub data82.4%
12Claude Fable 5max· Epoch AI Benchmarking Hub data81.9%
13Muse Spark 1.3· Epoch AI Benchmarking Hub data81.8%
14Claude Opus 5· Epoch AI Benchmarking Hub data80.6%
15Gemini 3.1 Pro Preview· Epoch AI Benchmarking Hub data79.6%
16GPT-6.2 Astrapro / max· Epoch AI Benchmarking Hub data79.2%
17GPT-6.2 Astra· Epoch AI Benchmarking Hub data77.3%
18GPT-5.5 Pro· Epoch AI Benchmarking Hub data76.9%
19Gemini 3.5 Flash· Epoch AI Benchmarking Hub data76.7%
20Gemini 3 Pro Preview· Epoch AI Benchmarking Hub data76.4%
21Claude Sonnet 5.5· Epoch AI Benchmarking Hub data75.9%
22Grok 4.6· Epoch AI Benchmarking Hub data75.9%
23Muse Spark 1.2· Epoch AI Benchmarking Hub data74.5%
24GPT-5.4 Pro· Epoch AI Benchmarking Hub data74.1%
25GPT-6 Sol· Epoch AI Benchmarking Hub data73.1%
26Muse Ember 1.0· Epoch AI Benchmarking Hub data72.7%
27GPT-6.1 Terra· Epoch AI Benchmarking Hub data72.2%
28GPT-5.6 Solpro· Epoch AI Benchmarking Hub data71.7%
29GPT-5.6 Solpro / xhigh· Epoch AI Benchmarking Hub data71.7%
30Qwen3.7 Max· Epoch AI Benchmarking Hub data70.4%
31Grok 4.5· Epoch AI Benchmarking Hub data70%
32GPT-6.2 Sol· Epoch AI Benchmarking Hub data69.9%
33GPT 5.5· Epoch AI Benchmarking Hub data69%
34Claude Opus 4.6· Epoch AI Benchmarking Hub data67.6%
35DeepSeek V4.1 Flash· Epoch AI Benchmarking Hub data66.7%
36GLM 5.3· Epoch AI Benchmarking Hub data66.2%
37Claude Opus 4.8· Epoch AI Benchmarking Hub data64.8%
38GPT-5.6 Sol· Epoch AI Benchmarking Hub data64.8%
39GPT-5.6 Solxhigh· Epoch AI Benchmarking Hub data64.8%
40Qwen3.6 Max Preview· Epoch AI Benchmarking Hub data63%
41Gemini 2.5 Pro· Epoch AI Benchmarking Hub data62.4%
42Claude Opus 4.5· Epoch AI Benchmarking Hub data62%
43Claude Opus 4.7· Epoch AI Benchmarking Hub data61.7%
44GPT-5 Pro· Epoch AI Benchmarking Hub data61.6%
45DeepSeek V4 Flash 0731· Epoch AI Benchmarking Hub data61.1%
46Gemini 3 Flash· Epoch AI Benchmarking Hub data61.1%
47Kimi K3max· Epoch AI Benchmarking Hub data60.7%
48Claude Sonnet 5· Epoch AI Benchmarking Hub data60.6%
49Grok 4· Epoch AI Benchmarking Hub data60.5%
50Qwen 3.8 27B· Epoch AI Benchmarking Hub data60.2%

Showing 1–50 of 114 results.

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