Skip to content

Benchmark

Epoch Capabilities Index

Models measured
197
Scores recorded
199
Sources
1
IQ weight
No weight published

Score distribution

137.4
Mean
142.4
Median
22.5
Std dev
109.1
10th percentile
160.2
90th percentile
55.5
Min
55.5173.4
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
51GPT 5.2· Epoch AI Benchmarking Hub data154.5
52DeepSeek V4 Flash 0731· Epoch AI Benchmarking Hub data154.4
53GLM 5.3 Flash· Epoch AI Benchmarking Hub data153.8
54Muse Spark· Epoch AI Benchmarking Hub data153.2
55GLM 5.2· Epoch AI Benchmarking Hub data153.1
56Gemini 3 Pro Preview· Epoch AI Benchmarking Hub data153
57Claude Sonnet 4.6· Epoch AI Benchmarking Hub data152.9
58Qwen 3.8 27B· Epoch AI Benchmarking Hub data152.6
59Gemini 3 Flash· Epoch AI Benchmarking Hub data152.4
60Grok 4.20· Epoch AI Benchmarking Hub data152.3
61Claude Opus 4.5· Epoch AI Benchmarking Hub data151.9
62Kimi K2.6· Epoch AI Benchmarking Hub data151.6
63GPT-5 Pro· Epoch AI Benchmarking Hub data151.3
64Kimi K2.7 Code· Epoch AI Benchmarking Hub data151.2
65GPT 5.1· Epoch AI Benchmarking Hub data150.9
66Inkling Small· Epoch AI Benchmarking Hub data150.8
67GLM 5.1· Epoch AI Benchmarking Hub data150.7
68DeepSeek V4 Pro· Epoch AI Benchmarking Hub data150.2
69GPT 5· Epoch AI Benchmarking Hub data150
70Qwen3.6 Max Preview· Epoch AI Benchmarking Hub data149.9
71GPT-5.4 mini· Epoch AI Benchmarking Hub data149.4
72Qwen3.6 Plus· Epoch AI Benchmarking Hub data148.9
73Kimi K2.5· Epoch AI Benchmarking Hub data148.8
74Inkling FP4· Epoch AI Benchmarking Hub data148.4
75Qwen3.5 397B A17B· Epoch AI Benchmarking Hub data148
76Qwen3.7 Plus· Epoch AI Benchmarking Hub data147.9
77MiniMax M3· Epoch AI Benchmarking Hub data147.8
78Claude Sonnet 4.5· Epoch AI Benchmarking Hub data147.6
79O3 Pro· Epoch AI Benchmarking Hub data147.5
80Nemotron 3 Ultra· Epoch AI Benchmarking Hub data147
81OpenAI o3· Epoch AI Benchmarking Hub data147
82DeepSeek V4 Flash· Epoch AI Benchmarking Hub data146.9
83Grok 4· Epoch AI Benchmarking Hub data146.9
84Qwen3.6 27B· Epoch AI Benchmarking Hub data146.7
85GLM 5· Epoch AI Benchmarking Hub data146.5
86GPT 5.4 Nano· Epoch AI Benchmarking Hub data146.4
87DeepSeek V3.2· Epoch AI Benchmarking Hub data146.3
88GPT-5 mini· Epoch AI Benchmarking Hub data146
89Gemini 3.5 Flash Lite· Epoch AI Benchmarking Hub data145.6
90Kimi K2 Thinking· Epoch AI Benchmarking Hub data145.5
91o4 Mini· Epoch AI Benchmarking Hub data145.2
92Gemini 3.1 Flash Lite· Epoch AI Benchmarking Hub data145
93Gemini 2.5 Pro· Epoch AI Benchmarking Hub data144.9
94Claude Opus 4.1· Epoch AI Benchmarking Hub data144.8
95DeepSeek V3.2 Exp· Epoch AI Benchmarking Hub data144.4
96Grok 4 Fast· Epoch AI Benchmarking Hub data144.3
97Gemma 4 31B IT· Epoch AI Benchmarking Hub data144.1
98Qwen3.6 35B A3B· Epoch AI Benchmarking Hub data144.1
99Qwen3.5 35B A3B· Epoch AI Benchmarking Hub data144.1
100Qwen3.6 Flash· Epoch AI Benchmarking Hub data143.7

Showing 51–100 of 199 results.

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