Noometry

Math benchmark

LiveBench Math leaderboard

As of October 2026, GPT-5.1 has the highest published LiveBench Math score on Noometry at 94.5%, out of 39 models with results.

Last verified

About LiveBench Math

LiveBench math tasks drawn from recent competitions.

Category
Math
Introduced
2024
Unit
Percent (random guessing ≈ 0%)
Official site
livebench.ai

Top 15 models

  1. GPT-5.1 94.5%
  2. Gemini 2.5 Pro 90.2%
  3. DeepSeek-R1 80.7%
  4. o1 80.3%
  5. Claude 3.7 Sonnet 79%
  6. QwQ-32B 77.8%
  7. o3-mini 77.3%
  8. Gemini 2.0 Flash (Feb 2025) 75.8%
  9. DeepSeek-V3 73.5%
  10. Gemini 2.0 Pro 71%
  11. GPT-4.5 69.3%
  12. o1-mini 62%
  13. DeepSeek-R1-Distill-Qwen-32B 59.4%
  14. Qwen2.5-Max 58.4%
  15. DeepSeek-R1-Distill-Llama-70B 58.1%

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

All results

#ModelProviderScoreSettingSourceDate
1GPT-5.1 OpenAI94.5%highEpoch AI
2Gemini 2.5 Pro Google90.2%Epoch AI
3DeepSeek-R1 DeepSeek80.7%Epoch AI
4o1 OpenAI80.3%highEpoch AI
5Claude 3.7 Sonnet Anthropic79%Epoch AI
6QwQ-32B Alibaba (Qwen)77.8%Epoch AI
7o3-mini OpenAI77.3%highEpoch AI
8Gemini 2.0 Flash (Feb 2025) Google75.8%Epoch AI
9DeepSeek-V3 DeepSeek73.5%Epoch AI
10Gemini 2.0 Pro Google71%Epoch AI
11GPT-4.5 OpenAI69.3%Epoch AI
12o1-mini OpenAI62%mediumEpoch AI
13DeepSeek-R1-Distill-Qwen-32B DeepSeek59.4%Epoch AI
14Qwen2.5-Max Alibaba (Qwen)58.4%Epoch AI
15DeepSeek-R1-Distill-Llama-70B DeepSeek58.1%Epoch AI
16Gemini 2.0 Flash-Lite Google58.1%Epoch AI
17Gemma 3 27B Google55.4%Epoch AI
18Grok-2 (Dec 2024) xAI54.9%Epoch AI
19Claude 3.5 Sonnet Anthropic52.3%Epoch AI
20GPT-4o OpenAI49.5%Epoch AI
21Qwen2.5-Coder-32B Alibaba (Qwen)46.6%Epoch AI
22Claude 3 Opus Anthropic43.6%Epoch AI
23Mistral Large Mistral AI42.5%Epoch AI
24Llama-3.3-70B-Instruct Meta42.2%Epoch AI
25Phi-4 Microsoft42%Epoch AI
26Sonar Perplexity41.6%Epoch AI
27Mistral Small Mistral AI39.9%Epoch AI
28Amazon Nova Pro Amazon38%Epoch AI
29Amazon Nova Lite Amazon36.7%Epoch AI
30GPT-4o mini OpenAI36.3%Epoch AI
31Claude 3.5 Haiku Anthropic35.5%Epoch AI
32Amazon Nova Micro Amazon34.5%Epoch AI
33Gemma 2 27B Google26.5%Epoch AI
34Command R+ Cohere21.3%Epoch AI
35Gemma 2 9B Google19.8%Epoch AI
36Command R Cohere19.4%Epoch AI
37Phi 3 Small 8k Instruct Microsoft17.6%Epoch AI
38Phi 3 Mini 4k Instruct Microsoft15.7%Epoch AI
39OLMo 2 Furious 13B Allen Institute for AI (Ai2)13.6%Epoch AI

Compare the leaders

Other math benchmarks

Frequently asked questions

What does LiveBench Math measure?

LiveBench math tasks drawn from recent competitions.

Which model has the highest LiveBench Math score?

As of October 2026, GPT-5.1 has the highest published LiveBench Math score on Noometry at 94.5%, out of 39 models with results.

What is the best open-weight model on LiveBench Math?

QwQ-32B has the highest LiveBench Math accuracy among open-weight models at 77.8%, ranking 6 of 39 overall.