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
51GPT-5.1 Codex Max· Epoch AI Benchmarking Hub data1,208.8
52GPT 5.1high· Epoch AI Benchmarking Hub data1,192.2
53Qwen3.7 Max· Epoch AI Benchmarking Hub data1,189.4
54GPT-5.4 minihigh· Epoch AI Benchmarking Hub data1,188.6
55Gemini 3 Pro Preview· Epoch AI Benchmarking Hub data1,176.8
56GPT 5high· Epoch AI Benchmarking Hub data1,162.5
57Gemini 3.1 Pro Preview· Epoch AI Benchmarking Hub data1,160.6
58MiMo V2.6 Pro· Epoch AI Benchmarking Hub data1,158.1
59Grok 4.20· Epoch AI Benchmarking Hub data1,150.3
60GPT 5.5none· Epoch AI Benchmarking Hub data1,127.6
61Kimi K2.6· Epoch AI Benchmarking Hub data1,092.7
62DeepSeek V4.1 Flashmax· Epoch AI Benchmarking Hub data1,092.4
63GPT 5.4none· Epoch AI Benchmarking Hub data1,086
64Gemini 3.1 Pro Previewlow· Epoch AI Benchmarking Hub data1,054.8
65GLM 5.2high· Epoch AI Benchmarking Hub data1,047
66Claude Opus 4.516k· Epoch AI Benchmarking Hub data1,025.4
67GLM 5.2max· Epoch AI Benchmarking Hub data1,010.2
68DeepSeek V4 Prohigh· Epoch AI Benchmarking Hub data1,006.1
69GPT 5.4 Nanohigh· Epoch AI Benchmarking Hub data1,004.5
70Claude Opus 4.6· Epoch AI Benchmarking Hub data996.5
71Gemini 3 Pro Previewlow· Epoch AI Benchmarking Hub data988.2
72Inkling FP4· Epoch AI Benchmarking Hub data946
73Grok 4.3· Epoch AI Benchmarking Hub data944.2
74OpenAI o3high· Epoch AI Benchmarking Hub data933.6
75Gemma 4 26B A4B· Epoch AI Benchmarking Hub data927.2
76Gemma 4 31B IT· Epoch AI Benchmarking Hub data925.5
77Gemini 3.5 Flashhigh· Epoch AI Benchmarking Hub data911
78Gemini 3.7 Flashhigh· Epoch AI Benchmarking Hub data904.3
79MiMo V2.5 Pro· Epoch AI Benchmarking Hub data899.8
80GLM 5.1· Epoch AI Benchmarking Hub data887.1
81Kimi K2.7 Code· Epoch AI Benchmarking Hub data886.2
82o4 Minihigh· Epoch AI Benchmarking Hub data826.2
83Kimi K2.5· Epoch AI Benchmarking Hub data821.7
84GPT 5minimal· Epoch AI Benchmarking Hub data807.7
85DeepSeek R1· Epoch AI Benchmarking Hub data804.1
86GPT-5 minihigh· Epoch AI Benchmarking Hub data799.8
87Gemini 3.1 Flash Lite· Epoch AI Benchmarking Hub data797.7
88Claude Sonnet 4.532k· Epoch AI Benchmarking Hub data796.2
89Mercury 2· Epoch AI Benchmarking Hub data785.6
90Gemini 2.5 Pro32k· Epoch AI Benchmarking Hub data785.5
91MiMo V2 Pro· Epoch AI Benchmarking Hub data785.2
92GLM 5· Epoch AI Benchmarking Hub data765.6
93Gemini 3.5 Flash Litehigh· Epoch AI Benchmarking Hub data765.3
94Mistral Medium2604· Epoch AI Benchmarking Hub data764
95DeepSeek V3.1 Terminus· Epoch AI Benchmarking Hub data745.2
96MiMo V2 Flash· Epoch AI Benchmarking Hub data738
97GPT-5 nanohigh· Epoch AI Benchmarking Hub data718.7
98Gemini 3.6 Flashhigh· Epoch AI Benchmarking Hub data715.5
99Step 3.7 Flash· Epoch AI Benchmarking Hub data694.1
100DeepSeek V4 Flashhigh· Epoch AI Benchmarking Hub data678.2

Showing 51–100 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)