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
1Claude Fable 5.2· Epoch AI Benchmarking Hub data98.7%
2Claude Fable 5.2xhigh· Epoch AI Benchmarking Hub data98.2%
3Claude Opus 5.5max· Epoch AI Benchmarking Hub data98.2%
4Claude Fable 5max· Epoch AI Benchmarking Hub data97.3%
5Gemini 4 Carbonmedium· Epoch AI Benchmarking Hub data97.3%
6Claude Fable 5.2high· Epoch AI Benchmarking Hub data96.9%
7Claude Fable 5.2low· Epoch AI Benchmarking Hub data96.9%
8Claude Fable 5.2medium· Epoch AI Benchmarking Hub data96.9%
9Gemini 4 Carbon· Epoch AI Benchmarking Hub data96.9%
10Claude Opus 5high· Epoch AI Benchmarking Hub data96.5%
11Claude Opus 5xhigh· Epoch AI Benchmarking Hub data96.5%
12Claude Fable 5.1· Epoch AI Benchmarking Hub data96%
13Claude Fable 5.1xhigh· Epoch AI Benchmarking Hub data96%
14Claude Opus 5max· Epoch AI Benchmarking Hub data96%
15Gemini 4 Argonlow· Epoch AI Benchmarking Hub data96%
16Gemini 4 Carbonlow· Epoch AI Benchmarking Hub data96%
17Claude Fable 5xhigh· Epoch AI Benchmarking Hub data95.6%
18Claude Opus 5.5xhigh· Epoch AI Benchmarking Hub data95.6%
19Claude Sonnet 5.5max· Epoch AI Benchmarking Hub data95.6%
20Gemini 4 Argonhigh· Epoch AI Benchmarking Hub data95.6%
21Gemini 4 Argonmedium· Epoch AI Benchmarking Hub data95.6%
22GPT-6 Astraxhigh· Epoch AI Benchmarking Hub data95.6%
23GPT-6 Solmax· Epoch AI Benchmarking Hub data95.6%
24Grok 4.6xhigh· Epoch AI Benchmarking Hub data95.6%
25Claude Fable 5.1medium· Epoch AI Benchmarking Hub data95.1%
26Gemini 3.1 Pro Previewmedium· Epoch AI Benchmarking Hub data95.1%
27GPT-6 Astra· Epoch AI Benchmarking Hub data95.1%
28GPT-6 Astramedium· Epoch AI Benchmarking Hub data95.1%
29Gemini 3.7 Flashhigh· Epoch AI Benchmarking Hub data94.7%
30Claude Fable 5high· Epoch AI Benchmarking Hub data94.2%
31Claude Fable 5.1high· Epoch AI Benchmarking Hub data94.2%
32Claude Opus 5.5high· Epoch AI Benchmarking Hub data94.2%
33Claude Opus 5.5medium· Epoch AI Benchmarking Hub data94.2%
34GPT-6 Astrahigh· Epoch AI Benchmarking Hub data94.2%
35GPT-6.1 Solmax· Epoch AI Benchmarking Hub data94.2%
36Grok 4.5high· Epoch AI Benchmarking Hub data94.2%
37Muse Spark 1.3· Epoch AI Benchmarking Hub data94.2%
38Claude Sonnet 5.5xhigh· Epoch AI Benchmarking Hub data93.3%
39GPT 5.5xhigh· Epoch AI Benchmarking Hub data93.3%
40GPT-5.5 Proxhigh· Epoch AI Benchmarking Hub data93.3%
41GPT-5.6 Solpro / max· Epoch AI Benchmarking Hub data93.3%
42Grok 4.7xhigh· Epoch AI Benchmarking Hub data93.3%
43Claude Fable 5medium· Epoch AI Benchmarking Hub data92.9%
44GPT-5.6 Solxhigh· Epoch AI Benchmarking Hub data92.9%
45Gemini 3.8 Flash· Epoch AI Benchmarking Hub data92.9%
46Claude Opus 5.5low· Epoch AI Benchmarking Hub data92.5%
47Gemini 3.1 Pro Previewhigh· Epoch AI Benchmarking Hub data92.5%
48Gemini 3.6 Flashhigh· Epoch AI Benchmarking Hub data92.5%
49GPT-5.6 Solhigh· Epoch AI Benchmarking Hub data92.5%
50GPT-5.6 Solmax· Epoch AI Benchmarking Hub data92.5%

Showing 1–50 of 239 results.

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