Skip to content

Benchmark

Blueprint Bench 2 Score

Scores are percentages.

Models measured
40
Scores recorded
40
Sources
1
IQ weight
3.46

Score distribution

30.9%
Mean
32.5%
Median
17%
Std dev
0%
10th percentile
54.4%
90th percentile
0%
Min
0%61.4%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 3.46 for chat models, measured over 40 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1GPT-6.2 Astra· Epoch AI Benchmarking Hub data61.4%
2Gemini 4 Carbon· Epoch AI Benchmarking Hub data58.6%
3GPT-6.2 Sol· Epoch AI Benchmarking Hub data57%
4Claude Fable 5.2· Epoch AI Benchmarking Hub data56.6%
5Gemini 4 Argon· Epoch AI Benchmarking Hub data54.4%
6Claude Opus 5.5· Epoch AI Benchmarking Hub data51.2%
7GPT-6 Astra· Epoch AI Benchmarking Hub data49.7%
8GPT-6.1 Sol· Epoch AI Benchmarking Hub data43.2%
9GPT-6.1 Terra· Epoch AI Benchmarking Hub data42.1%
10Claude Fable 5.1· Epoch AI Benchmarking Hub data41.9%
11Claude Sonnet 5.5· Epoch AI Benchmarking Hub data39.3%
12Claude Fable 5· Epoch AI Benchmarking Hub data38.6%
13Gemini 3.8 Flash· Epoch AI Benchmarking Hub data38.6%
14GPT-6 Sol· Epoch AI Benchmarking Hub data36.9%
15GPT 5.5· Epoch AI Benchmarking Hub data36.2%
16Gemini 3.5 Flash· Epoch AI Benchmarking Hub data33.6%
17GPT-5.6 Sol· Epoch AI Benchmarking Hub data33.6%
18Grok 4.6· Epoch AI Benchmarking Hub data33.2%
19Claude Haiku 5.5· Epoch AI Benchmarking Hub data33%
20Grok 4.7· Epoch AI Benchmarking Hub data32.5%
21Muse Wonder 1.0· Epoch AI Benchmarking Hub data31.7%
22Gemini 3.6 Flash· Epoch AI Benchmarking Hub data31.2%
23GPT-6 Luna· Epoch AI Benchmarking Hub data31.2%
24GPT-5.6 Terra· Epoch AI Benchmarking Hub data30.8%
25Claude Opus 5· Epoch AI Benchmarking Hub data30.4%
26Kimi K3· Epoch AI Benchmarking Hub data29.5%
27Grok 4.5· Epoch AI Benchmarking Hub data27.3%
28GPT 5.4· Epoch AI Benchmarking Hub data27.1%
29Gemini 3.1 Pro Preview· Epoch AI Benchmarking Hub data26.5%
30Claude Sonnet 5· Epoch AI Benchmarking Hub data24.9%
31Claude Opus 4.7· Epoch AI Benchmarking Hub data24.5%
32GPT-5.6 Luna· Epoch AI Benchmarking Hub data22.6%
33Claude Opus 4.8· Epoch AI Benchmarking Hub data14.5%
34Claude Sonnet 4.6· Epoch AI Benchmarking Hub data6.7%
35Kimi K2.6· Epoch AI Benchmarking Hub data3.9%
36Claude Haiku 4.5· Epoch AI Benchmarking Hub data0%
37Gemini 3 Flash· Epoch AI Benchmarking Hub data0%
38Gemini Robotics-ER 1.6 Preview· Epoch AI Benchmarking Hub data0%
39Grok 4.20· Epoch AI Benchmarking Hub data0%
40Grok 4.3· Epoch AI Benchmarking Hub data0%

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