Benchmark Explorer

Explore how models perform on various benchmarks

Benchmarks

🔍

LiveBench (Math)

Evaluates performance on recent high school math competition problems and enhanced AMPS questions from the past 12 months.
Source:

Model Performance

#1
GPT-5
92.8%
#2
Claude-4.1-Opus-Thinking
91.2%
#3
Claude-4.0-Opus-Thinking
88.2%
#4
GPT-5-mini
86.0%
#5
DeepSeek-R1
85.3%
#6
Claude-4.0-Sonnet-Thinking
85.2%
#7
OpenAI-O3-high
85.0%
#8
OpenAI-O4-mini-high
84.9%
#9
Gemini-2.5-Flash
84.1%
#10
Gemini-2.5-Pro-0325
83.3%
#11
OpenAI-O4-mini-medium
81.0%
#12
OpenAI-O3-medium
80.7%
#13
Qwen-3
80.0%
#14
Claude-3.7-Sonnet-Thinking
79.0%
#15
DeepSeek-V3
78.9%
#16
Claude-4.0-Opus
78.8%
#17
Grok-3-Mini-Beta
77.0%
#18
Claude-4.0-Sonnet
76.4%
#19
Kimi-K2-Instruct
74.4%
#20
Gemini-2.5-Flash-Thinking
71.1%
#21
GPT-OSS-120B
69.5%
#22
GPT-4.5
67.9%
#23
Claude-3.7-Sonnet
64.7%
#24
Phi-4
62.8%
#25
GPT-4.1
62.4%
#26
Llama-4-Maverick-17B
60.6%
#27
Mistral-Medium-3.1
59.7%
#28
GPT-5-nano
59.0%
#29
GPT-4o-0513
55.7%
#30
Claude-3.5-Sonnet-1022
50.5%
#31
Cohere-Command-A
45.5%
#32
Mistral-Large
42.2%
#33
Claude-3.5-Haiku
34.8%
#34
Cohere-Command-R-Plus
22.8%