Skip to content

Benchmark

Dtbench CRI-rescaled score

Scores are percentages.

Models measured
170
Scores recorded
239
Sources
1
IQ weight
1.22

Score distribution

66.3%
Mean
74.2%
Median
27.6%
Std dev
20.9%
10th percentile
95.1%
90th percentile
2.8%
Min
2.8%98.7%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.22 for chat models, measured over 170 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
51GPT-5.6 Solmedium· Epoch AI Benchmarking Hub data92%
52Muse Wonder 1.0· Epoch AI Benchmarking Hub data92%
53Claude Haiku 5.5max· Epoch AI Benchmarking Hub data91.7%
54Claude Opus 4.8max· Epoch AI Benchmarking Hub data91.6%
55Claude Opus 5medium· Epoch AI Benchmarking Hub data91.6%
56Gemini 3.1 Pro Previewlow· Epoch AI Benchmarking Hub data91.6%
57Gemini 3.8 Flashlow· Epoch AI Benchmarking Hub data91.6%
58Gemini 3.8 Flashmedium· Epoch AI Benchmarking Hub data91.6%
59GPT-6 Astralow· Epoch AI Benchmarking Hub data91.6%
60Gemini 3.5 Flashhigh· Epoch AI Benchmarking Hub data91.1%
61Gemini 3.7 Flashmedium· Epoch AI Benchmarking Hub data91.1%
62Muse Spark 1.2xhigh· Epoch AI Benchmarking Hub data91.1%
63Claude Opus 4.7max· Epoch AI Benchmarking Hub data91.1%
64GPT 5.4xhigh· Epoch AI Benchmarking Hub data90.7%
65Muse Spark 1.1high· Epoch AI Benchmarking Hub data90.7%
66Claude Fable 5.1low· Epoch AI Benchmarking Hub data90.2%
67Claude Sonnet 5.5high· Epoch AI Benchmarking Hub data90.2%
68Claude Sonnet 5.5medium· Epoch AI Benchmarking Hub data89.8%
69DeepSeek V4 Pro 0813· Epoch AI Benchmarking Hub data89.8%
70GPT-5.6 Sollow· Epoch AI Benchmarking Hub data89.8%
71GLM 5.2max· Epoch AI Benchmarking Hub data89.3%
72Claude Sonnet 5.5low· Epoch AI Benchmarking Hub data88.9%
73Gemini 3.7 Flashlow· Epoch AI Benchmarking Hub data88.9%
74GPT-5.6 Terramax· Epoch AI Benchmarking Hub data88.9%
75Kimi K3.1 Promax· Epoch AI Benchmarking Hub data88.9%
76Claude Opus 5low· Epoch AI Benchmarking Hub data88.5%
77GPT-5.6 Terraxhigh· Epoch AI Benchmarking Hub data87.6%
78Claude Sonnet 5max· Epoch AI Benchmarking Hub data87.5%
79Qwen3.7 Max· Epoch AI Benchmarking Hub data87.1%
80Qwen3.8 Maxxhigh· Epoch AI Benchmarking Hub data86.7%
81Claude Fable 5low· Epoch AI Benchmarking Hub data85.8%
82Claude Opus 4.6max· Epoch AI Benchmarking Hub data85.3%
83GPT-5.6 Terrahigh· Epoch AI Benchmarking Hub data85.3%
84Kimi K3max· Epoch AI Benchmarking Hub data85.3%
85DeepSeek V4 Flash 0731max· Epoch AI Benchmarking Hub data84.9%
86GPT 5.2xhigh· Epoch AI Benchmarking Hub data84.9%
87Kimi K2.6· Epoch AI Benchmarking Hub data84.9%
88DeepSeek V4 Promax· Epoch AI Benchmarking Hub data84.5%
89GPT 5high· Epoch AI Benchmarking Hub data84.5%
90Grok 4.3high· Epoch AI Benchmarking Hub data84.5%
91GPT 5.1high· Epoch AI Benchmarking Hub data83.6%
92GPT-5.6 Terralow· Epoch AI Benchmarking Hub data83.6%
93GPT-5.6 Terramedium· Epoch AI Benchmarking Hub data83.6%
94GPT-6 Lunamax· Epoch AI Benchmarking Hub data83.6%
95Grok 4.20· Epoch AI Benchmarking Hub data83.6%
96Nemotron 3 Ultra· Epoch AI Benchmarking Hub data83.6%
97Claude Opus 4.5high· Epoch AI Benchmarking Hub data83.1%
98Claude Sonnet 4.6max· Epoch AI Benchmarking Hub data83.1%
99DeepSeek V4.1 Flash· Epoch AI Benchmarking Hub data83.1%
100Claude Sonnet 5xhigh· Epoch AI Benchmarking Hub data82.5%

Showing 51–100 of 239 results.

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