Benchmark Explorer

Explore how models perform on various benchmarks

Benchmarks

🔍

LiveBench (Language)

Evaluates language comprehension through tasks including Connections word puzzles, typo removal, and movie synopsis unscrambling from recent sources.
Source:

Model Performance

#1
GPT-5
80.8%
#2
Claude-4.0-Opus
76.1%
#3
OpenAI-O3-high
76.0%
#4
Gemini-2.5-Pro-0325
74.5%
#5
Claude-4.0-Opus-Thinking
73.7%
#6
OpenAI-O3-medium
73.5%
#7
Claude-4.1-Opus-Thinking
71.2%
#8
Claude-4.0-Sonnet-Thinking
70.2%
#9
GPT-5-mini
68.8%
#10
Claude-3.7-Sonnet-Thinking
68.3%
#11
Claude-4.0-Sonnet
67.2%
#12
OpenAI-O4-mini-high
66.0%
#13
DeepSeek-R1
64.8%
#14
GPT-4.5
64.8%
#15
Kimi-K2-Instruct
63.9%
#16
Claude-3.7-Sonnet
63.2%
#17
OpenAI-O4-mini-medium
62.4%
#18
DeepSeek-V3
59.8%
#19
Grok-3-Mini-Beta
59.1%
#20
Gemini-2.5-Flash
57.0%
#21
Qwen-3
55.1%
#22
GPT-4.1
54.5%
#23
Claude-3.5-Sonnet-1022
54.5%
#24
Gemini-2.5-Flash-Thinking
52.6%
#25
GPT-5-nano
51.0%
#26
Llama-4-Maverick-17B
49.6%
#27
GPT-4o-0513
49.4%
#28
Mistral-Medium-3.1
44.7%
#29
GPT-OSS-120B
43.1%
#30
Mistral-Large
40.5%
#31
Claude-3.5-Haiku
39.7%
#32
Cohere-Command-A
36.7%
#33
Cohere-Command-R-Plus
30.9%
#34
Phi-4
30.7%