AIME 2025

Reasoning

Source ↗ Higher is better

Rank	Model	Provider	Score
1	GPT-5.2 (xhigh)	OpenAI	99.0%
2	GPT-5 Codex (high)	OpenAI	98.7%
3	Gemini 3 Flash Preview (Reasoning)	Google	97.0%
4	DeepSeek V3.2 Speciale	DeepSeek	96.7%
5	GPT-5.2 (medium)	OpenAI	96.7%
6	MiMo-V2-Flash (Reasoning)	Xiaomi	96.3%
7	Gemini 3 Pro Preview (high)	Google	95.7%
8	GPT-5.1 Codex (high)	OpenAI	95.7%
9	GLM-4.7 (Reasoning)	Z.ai	95.0%
10	KAT-Coder-Pro V1	Kwaipilot	94.7%
11	MoonshotAI: Kimi K2 Thinking	Moonshot AI	94.7%
12	GPT-5 (high)	OpenAI	94.3%
13	Nova 2.0 Lite (high)	Amazon	94.3%
14	GPT-5.1 (high)	OpenAI	94.0%
15	gpt-oss-120b (high)	OpenAI	93.4%
16	Grok 4	SpaceXAI	92.7%
17	DeepSeek V3.2 (Reasoning)	DeepSeek	92.0%
18	GPT-5 (medium)	OpenAI	91.7%
19	GPT-5.1 Codex mini (high)	OpenAI	91.7%
20	Claude Opus 4.5 (Reasoning)	Anthropic	91.3%
21	NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)	NVIDIA	91.0%
22	Qwen3 235B A22B 2507 (Reasoning)	Alibaba	91.0%
23	OpenAI: o4 Mini High	OpenAI	90.7%
24	GPT-5 mini (high)	OpenAI	90.7%
25	K-EXAONE (Reasoning)	LG AI Research	90.3%
26	grok-4-fast-reasoning	SpaceXAI	89.7%
27	DeepSeek V3.1 (Reasoning)	DeepSeek	89.7%
28	Nova 2.0 Omni (medium)	Amazon	89.7%
29	DeepSeek V3.1 Terminus (Reasoning)	DeepSeek	89.7%
30	grok-4-1-fast-reasoning	SpaceXAI	89.3%
31	Ring-1T	inclusionAI	89.3%
32	gpt-oss-20b (high)	OpenAI	89.3%
33	Nova 2.0 Pro Preview (medium)	Amazon	89.0%
34	Nova 2.0 Lite (medium)	Amazon	88.7%
35	OpenAI: o3	OpenAI	88.3%
36	Qwen3 VL 235B A22B (Reasoning)	Alibaba	88.3%
37	Apriel-v1.6-15B-Thinker	ServiceNow	88.0%
38	intellect-3	Prime Intellect	88.0%
39	Claude 4.5 Sonnet (Reasoning)	Anthropic	88.0%
40	DeepSeek V3.2 Exp (Reasoning)	DeepSeek	87.7%
41	Google: Gemini 2.5 ProRetiring	Google	87.7%
42	Apriel-v1.5-15B-Thinker	ServiceNow	87.5%
43	Gemini 3 Pro Preview (low)	Google	86.7%
44	GLM-4.6 (Reasoning)	Z.ai	86.0%
45	GLM-4.6V (Reasoning)	Z.ai	85.3%
46	GPT-5 mini (medium)	OpenAI	85.0%
47	ERNIE 5.0 Thinking Preview	Baidu	85.0%
48	Qwen3 VL 32B (Reasoning)	Alibaba	84.7%
49	Seed-OSS-36B-Instruct	ByteDance Seed	84.7%
50	Grok 3 mini Reasoning (high)	SpaceXAI	84.7%

Access this data

JSON GET /api/v1

[
  {
    "rank": 1,
    "model": "GPT-5.2 (xhigh)",
    "provider": "OpenAI",
    "score": "99.0"
  },
  {
    "rank": 2,
    "model": "GPT-5 Codex (high)",
    "provider": "OpenAI",
    "score": "98.6666666666667"
  },
  {
    "rank": 3,
    "model": "Gemini 3 Flash Preview (Reasoning)",
    "provider": "Google",
    "score": "97.0"
  }
]

Curl

$ curl -s https://rightrank.ai/api/v1/rankings/aime-2025.json | jq '.[:3]'

CLI brew install rightrank

$ rightrank top aime-2025 --limit 5
$ rightrank compare claude-opus-4.6 gpt-5.4-pro --category aime-2025
$ rightrank value aime-2025  # best performance per dollar

Claude Code Skill Skill

# In Claude Code, use the RightRank skill:
/rightrank best model for aime-2025
/rightrank compare top 5 by value
/rightrank cheapest model scoring above 80 on aime-2025