API
Skills/Math & Reasoning

Math & Reasoning

Mathematical problem solving, logical reasoning, and multi-step inference β€” from arithmetic to competition-level mathematics.

165
Models Ranked
91.1
Top Score
35.0
Average Score
9
Benchmarks
1Alibaba QwenAlibaba Qwen91.1
2Alibaba QwenAlibaba Qwen86.7
3AnthropicAnthropic86.7
4Alibaba QwenAlibaba Qwen85.2
5GoogleGoogle83.5
6MicrosoftMicrosoft72.1
7DeepSeekDeepSeek71.7
8Google DeepMindGoogle DeepMind66.6
9AlibabaAlibaba65.8
10
U
unknown
64.3
11AnthropicAnthropic63.7
12MicrosoftMicrosoft62.3
13OpenAIOpenAI61.0
14OpenAIOpenAI59.9
15
U
unknown
57.7
16OpenAIOpenAI54.9
17OpenAIOpenAI54.9
18
T
TII
54.4
19GoogleGoogle54.3
20
T
TII
53.8
21Google DeepMindGoogle DeepMind52.7
22OpenAIOpenAI51.1
23OpenAIOpenAI51.1
24AlibabaAlibaba50.6
25MetaMeta49.5
26Alibaba QwenAlibaba Qwen48.1
27MistralMistral47.9
28
U
unknown
47.8
29OpenAIOpenAI47.7
30OpenAIOpenAI47.7
31OpenAIOpenAI47.7
32OpenAIOpenAI47.7
33Mistral AIMistral AI47.5
34OpenAIOpenAI47.2
35AnthropicAnthropic47.2
36Google DeepMindGoogle DeepMind46.8
37Google DeepMindGoogle DeepMind46.8
38Google DeepMindGoogle DeepMind46.8
39Google DeepMindGoogle DeepMind46.8
40Google DeepMindGoogle DeepMind46.8
41Google DeepMindGoogle DeepMind46.8
42AnthropicAnthropic46.7
43MicrosoftMicrosoft46.4
44MicrosoftMicrosoft45.9
45
U
unknown
45.5
46OpenAIOpenAI45.2
47OpenAIOpenAI45.2
48OpenAIOpenAI45.2
49OpenAIOpenAI45.2
50OpenAIOpenAI44.8