Noometry

Multimodal benchmark

Blueprint-Bench 2 leaderboard

As of October 2026, Gemini 4 Argon has the highest published Blueprint-Bench 2 score on Noometry at 54.4%, out of 31 models with results.

Last verified

About Blueprint-Bench 2

A description with primary sources is being prepared for this benchmark.

Category
Multimodal
Introduced
2026
Unit
Percent (random guessing ≈ 0%)
Official site
epoch.ai

Top 15 models

  1. Gemini 4 Argon 54.4%
  2. Claude Opus 5.5 51.2%
  3. GPT-6 Astra 49.7%
  4. Claude Fable 5.1 41.9%
  5. Claude Fable 5 38.6%
  6. Gemini 3.8 Flash 38.6%
  7. GPT-6 Sol 36.9%
  8. GPT-5.5 36.2%
  9. Gemini 3.5 Flash 33.6%
  10. GPT-5.6 Sol 33.6%
  11. Grok 4.6 33.2%
  12. Grok 4.7 32.5%
  13. Gemini 3.6 Flash 31.2%
  14. GPT-6 Luna 31.2%
  15. GPT-5.6 Terra 30.8%

Sponsored placements are available on pages like this one. Advertise on Noometry

All results

#ModelProviderScoreSettingSourceDate
1Gemini 4 Argon Google54.4%Epoch AI
2Claude Opus 5.5 Anthropic51.2%Epoch AI
3GPT-6 Astra OpenAI49.7%Epoch AI
4Claude Fable 5.1 Anthropic41.9%Epoch AI
5Claude Fable 5 Anthropic38.6%Epoch AI
6Gemini 3.8 Flash Google38.6%Epoch AI
7GPT-6 Sol OpenAI36.9%Epoch AI
8GPT-5.5 OpenAI36.2%Epoch AI
9Gemini 3.5 Flash Google33.6%Epoch AI
10GPT-5.6 Sol OpenAI33.6%Epoch AI
11Grok 4.6 xAI33.2%Epoch AI
12Grok 4.7 xAI32.5%Epoch AI
13Gemini 3.6 Flash Google31.2%Epoch AI
14GPT-6 Luna OpenAI31.2%Epoch AI
15GPT-5.6 Terra OpenAI30.8%Epoch AI
16Claude Opus 5 Anthropic30.4%Epoch AI
17Kimi K3 Moonshot AI29.5%Epoch AI
18Grok 4.5 xAI27.3%Epoch AI
19GPT-5.4 OpenAI27.1%Epoch AI
20Gemini 3.1 Pro Preview Google26.5%Epoch AI
21Claude Sonnet 5 Anthropic24.9%Epoch AI
22Claude Opus 4.7 Anthropic24.5%Epoch AI
23GPT-5.6 Luna OpenAI22.6%Epoch AI
24Claude Opus 4.8 Anthropic14.5%Epoch AI
25Claude Sonnet 4.6 Anthropic6.7%Epoch AI
26Kimi K2.6 Moonshot AI3.9%Epoch AI
27Claude Haiku 4.5 Anthropic0%Epoch AI
28Gemini 3 Flash Preview Google0%Epoch AI
29Gemini Robotics-ER 1.6 Google0%Epoch AI
30Grok 4.20 (Non-Reasoning) xAI0%Epoch AI
31Grok 4.3 xAI0%Epoch AI

Compare the leaders

Other multimodal benchmarks

Frequently asked questions

Which model has the highest Blueprint-Bench 2 score?

As of October 2026, Gemini 4 Argon has the highest published Blueprint-Bench 2 score on Noometry at 54.4%, out of 31 models with results.

What is the best open-weight model on Blueprint-Bench 2?

Kimi K3 has the highest Blueprint-Bench 2 accuracy among open-weight models at 29.5%, ranking 17 of 31 overall.