MATH

MATH dataset contains 12,500 challenging competition mathematics problems from AMC 10, AMC 12, AIME, and other mathematics competitions. Each problem includes full step-by-step solutions and spans multiple difficulty levels (1-5) across seven mathematical subjects including Prealgebra, Algebra, Number Theory, Counting and Probability, Geometry, Intermediate Algebra, and Precalculus.

id math · max 1 · 71 models reported

#	Model	Score
1	o3-mini OpenAI	0.98
2	o1 OpenAI	0.96
3	MiniStral 3 (14B Instruct 2512) Mistral AI · open	0.90
4	Mistral Large 3 Mistral AI · open	0.90
5	Gemini 2.0 Flash Google	0.90
6	Kimi K2 0905 Moonshot AI	0.89
7	Gemma 3 27B Google · open	0.89
8	Ministral 3 (8B Instruct 2512) Mistral AI · open	0.88
9	Gemini 2.0 Flash-Lite Google	0.87
10	Gemini 1.5 Pro Google	0.86
11	MiMo-V2.5-Pro Xiaomi · open	0.86
12	o1-preview OpenAI	0.85
13	GPT-5 OpenAI	0.85
14	Gemma 3 12B Google · open	0.84
15	Qwen2.5 32B Instruct Alibaba Cloud / Qwen Team · open	0.83
16	Qwen2.5 72B Instruct Alibaba Cloud / Qwen Team · open	0.83
17	Ministral 3 (3B Instruct 2512) Mistral AI · open	0.83
18	Qwen2.5 VL 32B Instruct Alibaba Cloud / Qwen Team · open	0.82
19	Phi 4 Microsoft · open	0.80
20	Qwen2.5 14B Instruct Alibaba Cloud / Qwen Team · open	0.80
21	Claude 3.5 Sonnet Anthropic	0.78
22	Gemini 1.5 Flash Google	0.78
23	Llama 3.3 70B Instruct Meta · open	0.77
24	GPT-4o OpenAI	0.77
25	Nova Pro Amazon	0.77
26	Grok-2 xAI	0.76
27	Gemma 3 4B Google · open	0.76
28	Qwen2.5 7B Instruct Alibaba Cloud / Qwen Team · open	0.76
29	DeepSeek-V2.5 DeepSeek · open	0.75
30	Llama 3.1 405B Instruct Meta · open	0.74
31	Nova Lite Amazon	0.73
32	Grok-2 mini xAI	0.73
33	GPT-4 Turbo OpenAI	0.73
34	Qwen3 235B A22B Alibaba Cloud / Qwen Team · open	0.72
35	Qwen2.5-Omni-7B Alibaba Cloud / Qwen Team · open	0.71
36	Claude 3.5 Sonnet Anthropic	0.71
37	Mistral Small 3 24B Instruct Mistral AI · open	0.71
38	GPT-4o mini OpenAI	0.70
39	Kimi K2 Base Moonshot AI · open	0.70
40	Mistral Small 3.2 24B Instruct Mistral AI · open	0.69
41	Claude 3.5 Haiku Anthropic	0.69
42	Mistral Small 3.1 24B Instruct Mistral AI · open	0.69
43	Nova Micro Amazon	0.69
44	Llama 3.2 90B Instruct Meta · open	0.68
45	Phi 4 Mini Microsoft · open	0.64
46	Llama 4 Maverick Meta · open	0.61
47	Claude 3 Opus Anthropic	0.60
48	Qwen2 72B Instruct Alibaba Cloud / Qwen Team · open	0.60
49	Phi-3.5-MoE-instruct Microsoft · open	0.59
50	Gemini 1.5 Flash 8B Google	0.59
51	Qwen2.5-Coder 32B Instruct Alibaba Cloud / Qwen Team · open	0.57
52	Ministral 8B Instruct Mistral AI · open	0.55
53	Llama 3.2 11B Instruct Meta · open	0.52
54	Grok-1.5 xAI	0.51
55	Llama 4 Scout Meta · open	0.50
56	Qwen2 7B Instruct Alibaba Cloud / Qwen Team · open	0.50
57	Phi-3.5-mini-instruct Microsoft · open	0.48
58	Pixtral-12B Mistral AI · open	0.48
59	Gemma 3 1B Google · open	0.48
60	Llama 3.2 3B Instruct Meta · open	0.48
61	Qwen2.5-Coder 7B Instruct Alibaba Cloud / Qwen Team · open	0.47
62	Mistral Small 3 24B Base Mistral AI · open	0.46
63	Claude 3 Sonnet Anthropic	0.43
64	GPT-3.5 Turbo OpenAI	0.43
65	Gemma 2 27B Google · open	0.42
66	GPT-4 OpenAI	0.42
67	Claude 3 Haiku Anthropic	0.39
68	Gemma 2 9B Google · open	0.37
69	Gemini 1.0 Pro Google	0.33
70	Hermes 3 70B Nous Research · open	0.21
71	ERNIE 4.5 Baidu	0.12