Benchmark Explorer

Explore how models perform on various benchmarks

Benchmarks

🔍

Humanity's Last Exam

A comprehensive benchmark developed by the Center for AI Safety and Scale AI to assess expert-level reasoning and knowledge across diverse fields. The dataset contains 3,000 questions from nearly 1,000 subject-matter experts across 500+ institutions and 50 countries, with 10% requiring image and text comprehension.
Source:

Model Performance

#1
GPT-5
27.0%
#2
GPT-5-Thinking
27.0%
#3
Grok-4-Thinking
24.0%
#4
Grok-4
24.0%
#5
Gemini-2.5-Pro-Thinking
21.0%
#6
GPT-5-mini-Thinking
20.0%
#7
OpenAI-O3-high
20.0%
#8
OpenAI-O3-medium
20.0%
#9
GPT-OSS-120B
19.0%
#10
OpenAI-O4-mini-high
18.0%
#11
OpenAI-O4-mini-medium
18.0%
#12
Gemini-2.5-Pro-0325
17.0%
#13
DeepSeek-R1
15.0%
#14
GPT-5-mini
15.0%
#15
Qwen-3-Thinking
15.0%
#16
Gemini-2.5-Pro-0605
15.0%
#17
Claude-4.1-Opus-Thinking
12.0%
#18
Claude-4.0-Opus-Thinking
12.0%
#19
OpenAI-O3-mini-high
12.0%
#20
Qwen-3
11.0%
#21
Grok-3-Mini-Beta
11.0%
#22
Gemini-2.5-Flash-Thinking
11.0%
#23
Claude-3.7-Sonnet-Thinking
10.0%
#24
Magistral-Medium-3.1
10.0%
#25
Claude-4.0-Sonnet-Thinking
10.0%
#26
Qwen-3-Max-Preview
9.0%
#27
OpenAI-O3-mini-medium
9.0%
#28
GPT-5-nano
8.0%
#29
OpenAI-O1-1217
8.0%
#30
GPT-5-nano-Thinking
8.0%
#31
Gemini-2.0-Flash-Thinking-0121
7.0%
#32
Gemini-2.0-Pro-0121
7.0%
#33
Kimi-K2-Instruct
7.0%
#34
Llama-2-7B
6.0%
#35
Claude-4.0-Opus
6.0%
#36
Llama-2-13B
5.0%
#37
Gemini-2.5-Flash
5.0%
#38
Grok-3-Beta
5.0%
#39
Claude-3.7-Sonnet
5.0%
#40
Cohere-Command-A
5.0%
#41
Cohere-Command-R-Plus
5.0%
#42
Gemini-1.5-Pro
5.0%
#43
GPT-4.1
5.0%
#44
Gemini-2.0-Flash
5.0%
#45
DeepSeek-V3-0324
5.0%
#46
Llama-2-70B
5.0%
#47
Llama-4-Maverick-17B
5.0%
#48
OpenAI-O1-mini
5.0%
#49
DeepSeek-V3
4.0%
#50
GPT-4-mini
4.0%
#51
Claude-3-Sonnet
4.0%
#52
Claude-3.5-Haiku
4.0%
#53
Claude-4.0-Sonnet
4.0%
#54
Claude-3.5-Sonnet-1022
4.0%
#55
Mistral-Large-2
4.0%
#56
Mistral-Medium-3.1
4.0%
#57
Llama-3.3-70B
4.0%
#58
Llama-3.1-405B
4.0%
#59
Phi-4
4.0%
#60
Claude-3-Opus
3.0%
#61
GPT-4o-0513
3.0%
#62
Mistral-Large
3.0%