Skip to content

Benchmark

ALE-Bench

Coding benchmarks compare models on programming tasks.

Models measured
128
Scores recorded
142
Sources
1
IQ weight
2.25

Score distribution

1,065.1
Mean
988.2
Median
593.5
Std dev
340.8
10th percentile
1,932.8
90th percentile
137.8
Min
137.82,951.3
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.25 for chat models, measured over 128 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#Model
1GPT-6 Astramax· Epoch AI Benchmarking Hub data2,951.3
2GPT-6.1 Solmax· Epoch AI Benchmarking Hub data2,743.8
3GPT-6.1 Terramax· Epoch AI Benchmarking Hub data2,608.8
4Claude Fable 5.2high· Epoch AI Benchmarking Hub data2,509.8
5GPT-6 Solmax· Epoch AI Benchmarking Hub data2,462
6Gemini 4 Carbonhigh· Epoch AI Benchmarking Hub data2,374.7
7GPT-5.6 Solmax· Epoch AI Benchmarking Hub data2,176.9
8Claude Opus 5high· Epoch AI Benchmarking Hub data2,164.6
9Claude Opus 5.5high· Epoch AI Benchmarking Hub data2,147.4
10Claude Fable 5.1high· Epoch AI Benchmarking Hub data2,143.2
11Gemini 4 Argonhigh· Epoch AI Benchmarking Hub data2,100.1
12Claude Fable 5high· Epoch AI Benchmarking Hub data2,041.3
13GPT-5.6 Terramax· Epoch AI Benchmarking Hub data1,951.4
14GPT 5.5xhigh· Epoch AI Benchmarking Hub data1,943
15Kimi K3.1 Promax· Epoch AI Benchmarking Hub data1,932.8
16Gemini 4 Hydrogenxhigh· Epoch AI Benchmarking Hub data1,825.7
17Claude Sonnet 5.5high· Epoch AI Benchmarking Hub data1,819.1
18Gemini 4 Argonlow· Epoch AI Benchmarking Hub data1,705.4
19Muse Spark 1.3xhigh· Epoch AI Benchmarking Hub data1,688.4
20Qwen3.8 Maxxhigh· Epoch AI Benchmarking Hub data1,679.1
21GPT-5.6 Lunamax· Epoch AI Benchmarking Hub data1,667.4
22GPT-5.3 Codexxhigh· Epoch AI Benchmarking Hub data1,655.2
23DeepSeek V4.1 Flash 1017max· Epoch AI Benchmarking Hub data1,622
24GPT 5.4high· Epoch AI Benchmarking Hub data1,607
25Muse Spark Open 816B A42Bxhigh· Epoch AI Benchmarking Hub data1,590.1
26GPT 5.5medium· Epoch AI Benchmarking Hub data1,589.4
27Claude Haiku 5.5high· Epoch AI Benchmarking Hub data1,587
28GPT-6 Lunaxhigh· Epoch AI Benchmarking Hub data1,576.9
29Claude Opus 4.8high· Epoch AI Benchmarking Hub data1,563.8
30Grok 4.7xhigh· Epoch AI Benchmarking Hub data1,545.9
31Kimi K3max· Epoch AI Benchmarking Hub data1,524.5
32GPT 5.4medium· Epoch AI Benchmarking Hub data1,520.7
33Grok 4.6xhigh· Epoch AI Benchmarking Hub data1,508.1
34Claude Sonnet 5high· Epoch AI Benchmarking Hub data1,463.1
35Claude Opus 4.8none· Epoch AI Benchmarking Hub data1,411.8
36DeepSeek V4 Pro 0813max· Epoch AI Benchmarking Hub data1,403.2
37Gemini 3 Flash· Epoch AI Benchmarking Hub data1,367.2
38Claude Sonnet 4.6medium· Epoch AI Benchmarking Hub data1,327.3
39Claude Opus 4.7· Epoch AI Benchmarking Hub data1,323.1
40GLM 5.3high· Epoch AI Benchmarking Hub data1,317.4
41Grok 4.5high· Epoch AI Benchmarking Hub data1,309.3
42DeepSeek V4 Flash 0731max· Epoch AI Benchmarking Hub data1,306.1
43GPT-5.2 Codex· Epoch AI Benchmarking Hub data1,299.9
44GPT 5.2high· Epoch AI Benchmarking Hub data1,293.6
45Gemini 3.8 Flashhigh· Epoch AI Benchmarking Hub data1,269.7
46GPT 5.2medium· Epoch AI Benchmarking Hub data1,249.8
47GPT-5.1 Codex· Epoch AI Benchmarking Hub data1,244.9
48MiMo V2.6 Flash· Epoch AI Benchmarking Hub data1,244.2
49GPT-5.1 Codex Maxxhigh· Epoch AI Benchmarking Hub data1,228.3
50Mistral Large 4· Epoch AI Benchmarking Hub data1,213.2

Showing 1–50 of 142 results.

Tokens the model generated per task, as measured by the source. Comparable within this benchmark only.

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