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
51Claude Opus 4.1· Epoch AI Benchmarking Hub data60%
52Claude Haiku 5.5· Epoch AI Benchmarking Hub data59.2%
53Claude Opus 4· Epoch AI Benchmarking Hub data58.8%
54GLM 5.2· Epoch AI Benchmarking Hub data58.8%
55Kimi K2.7 Code· Epoch AI Benchmarking Hub data57.9%
56GPT-5.2 Pro· Epoch AI Benchmarking Hub data57.4%
57GPT 5high· Epoch AI Benchmarking Hub data56.7%
58Grok 4.1 Fastnon-reasoning· Epoch AI Benchmarking Hub data56%
59Grok 4.1 Fastreasoning· Epoch AI Benchmarking Hub data56%
60GLM 5.1· Epoch AI Benchmarking Hub data55.1%
61Claude Sonnet 4.5· Epoch AI Benchmarking Hub data54.3%
62GLM 5· Epoch AI Benchmarking Hub data53.2%
63GPT 5.1high· Epoch AI Benchmarking Hub data53.2%
64OpenAI o3high· Epoch AI Benchmarking Hub data53.1%
65DeepSeek V3.2 Speciale· Epoch AI Benchmarking Hub data52.6%
66Gemini 2.5 Pro· Epoch AI Benchmarking Hub data51.6%
67Gemini 2.5 Pro Preview 03-25· Epoch AI Benchmarking Hub data51.6%
68Inkling FP4· Epoch AI Benchmarking Hub data50%
69GPT-5.6 Terra· Epoch AI Benchmarking Hub data48.9%
70GLM 4.7· Epoch AI Benchmarking Hub data47.7%
71GPT-5.6 Luna· Epoch AI Benchmarking Hub data46.8%
72Kimi K2.5· Epoch AI Benchmarking Hub data46.8%
73Claude Sonnet 3.7· Epoch AI Benchmarking Hub data46.4%
74Claude Sonnet 3.712k· Epoch AI Benchmarking Hub data46.4%
75DeepSeek V4 Flash· Epoch AI Benchmarking Hub data46.3%
76GPT 5.2· Epoch AI Benchmarking Hub data45.8%
77MiniMax M3· Epoch AI Benchmarking Hub data45.8%
78Claude Sonnet 4· Epoch AI Benchmarking Hub data45.5%
79Nemotron 3 Ultra· Epoch AI Benchmarking Hub data41.7%
80O1 Preview· Epoch AI Benchmarking Hub data41.7%
81Claude 3.5 Sonnet· Epoch AI Benchmarking Hub data41.4%
82Gemini 2.5 Flash· Epoch AI Benchmarking Hub data41.2%
83DeepSeek R1· Epoch AI Benchmarking Hub data40.8%
84o1high· Epoch AI Benchmarking Hub data40.1%
85DeepSeek V3.1· Epoch AI Benchmarking Hub data40%
86Kimi K2 Thinking· Epoch AI Benchmarking Hub data39.6%
87o4 Minihigh· Epoch AI Benchmarking Hub data38.7%
88o1medium· Epoch AI Benchmarking Hub data36.7%
89Grok 3· Epoch AI Benchmarking Hub data36.1%
90Qwen3.6 Flash· Epoch AI Benchmarking Hub data35.2%
91MiniMax M2.1· Epoch AI Benchmarking Hub data34.7%
92GPT-4.5 Preview· Epoch AI Benchmarking Hub data34.5%
93gemini-exp· Epoch AI Benchmarking Hub data31.1%
94Qwen 3 235B A22B· Epoch AI Benchmarking Hub data31%
95Gemini 2.0 Flash Experimentalthinking· Epoch AI Benchmarking Hub data30.7%
96Llama 4 Maverick· Epoch AI Benchmarking Hub data27.7%
97DeepSeek V3· Epoch AI Benchmarking Hub data27.2%
98gemini-1.5-pro-002· Epoch AI Benchmarking Hub data27.1%
99GPT 4.1· Epoch AI Benchmarking Hub data27%
100Kimi K2 Instruct· Epoch AI Benchmarking Hub data26.3%

Showing 51–100 of 114 results.

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