Math

AIME-2025

American Invitational Mathematics Examination 2025.

34 models published a score

#	Model	Company	Score
1	GPT-5.2	OpenAI	100.0
2	Grok 4 Heavy	xAI	100.0
3	Qwen3-Max-Thinking	Alibaba	100.0
4	Gemini 3 Flash	Google DeepMind	99.7
5	Doubao Seed 2.0 Pro	ByteDance	98.3
6	Step 3.5 Flash	StepFun	97.3
7	Kimi K2.5	Moonshot AI	96.1
8	DeepSeek V3.2 Speciale	DeepSeek	96.0
9	GLM-4.7	Zhipu AI	95.7
10	Gemini 3 Pro	Google DeepMind	95.0
11	DeepSeek V3.2	DeepSeek	93.1
12	Doubao Seed 2.0 Lite	ByteDance	93.0
13	EXAONE 4.5 33B	LG AI Research	92.9
14	K-EXAONE 236B-A23B	LG AI Research	92.8
15	Qwen3.6-35B-A3B	Alibaba	92.7
16	GLM-5	Zhipu AI	92.7
17	Qwen3.5-397B-A17B	Alibaba	91.3
18	Gemini 3.1 Pro	Google DeepMind	91.2
19	Nova 2 Lite	Amazon	91.0
20	Nemotron 3 Super	Nvidia	90.2
21	Gemma 4 (31B dense)	Google DeepMind	89.2
22	Nemotron 3 Nano	Nvidia	89.1
23	Gemma 4 26B-A4B	Google DeepMind	88.3
24	DeepSeek V4 Pro	DeepSeek	87.5
25	DeepSeek R1 0528	DeepSeek	87.5
26	MiniMax M2.5	MiniMax	86.3
27	Step-3	StepFun	82.9
28	GPT-5.5 Instant	OpenAI	81.2
29	Qwen3-Max	Alibaba	80.6
30	Magistral Medium 1.2	Mistral AI	65.0
31	Gemma 4 E4B	Google DeepMind	42.5
32	Mistral Large 3	Mistral AI	40.0
33	Gemma 4 E2B	Google DeepMind	37.5
34	Reka Flash 3	Reka	33.7