AIME

Reasoning

American Invitational Mathematics Examination problems

Higher is better

Rank	Model	Provider	Score
1	GPT-5 (high)	OpenAI	95.7%
2	Grok 4	SpaceXAI	94.3%
3	OpenAI: o4 Mini High	OpenAI	94.0%
4	Qwen3 235B A22B 2507 (Reasoning)	Alibaba	94.0%
5	Grok 3 mini Reasoning (high)	SpaceXAI	93.3%
6	GPT-5 (medium)	OpenAI	91.7%
7	Qwen3 30B A3B 2507 (Reasoning)	Alibaba	90.7%
8	OpenAI: o3	OpenAI	90.3%
9	DeepSeek R1 0528 (May '25)	DeepSeek	89.3%
10	Google: Gemini 2.5 ProRetiring	Google	88.7%
11	GLM-4.5 (Reasoning)	Z.ai	87.3%
12	Gemini 2.5 Pro Preview (Mar' 25)	Google	87.0%
14	Llama Nemotron Super 49B v1.5 (Reasoning)	NVIDIA	86.0%
15	MiniMax M1 80k	MiniMax	84.7%
16	Gemini 2.5 Pro Preview (May' 25)	Google	84.3%
17	EXAONE 4.0 32B (Reasoning)	LG AI Research	84.3%
18	Gemini 2.5 Flash Preview (Reasoning)	Google	84.3%
19	Qwen3 235B A22B (Reasoning)	Alibaba	84.0%
20	GPT-5 (low)	OpenAI	83.0%
21	Gemini 2.5 Flash (Reasoning)	Google	82.3%
22	MiniMax M1 40k	MiniMax	81.3%
23	Qwen3 32B (Reasoning)	Alibaba	80.7%
24	Perplexity: Sonar Reasoning Pro	Perplexity	79.0%
25	Qwen/QwQ-32B	Alibaba	78.0%
26	Claude 4 Sonnet (Reasoning)	Anthropic	77.3%
27	Sonar Reasoning	Perplexity	77.0%
28	OpenAI: o3 Mini	OpenAI	77.0%
13	↳OpenAI: o3 Mini High	OpenAI	86.0%
29	Qwen3 14B (Reasoning)	Alibaba	76.3%
30	Claude 4 Opus (Reasoning)	Anthropic	75.7%
31	Qwen3 30B A3B (Reasoning)	Alibaba	75.3%
32	Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)	NVIDIA	74.7%
33	Qwen3 8B (Reasoning)	Alibaba	74.7%
34	Qwen3 30B A3B 2507 Instruct	Alibaba	72.7%
35	OpenAI: o1	OpenAI	72.3%
36	Qwen3 235B A22B 2507 Instruct	Alibaba	71.7%
37	Magistral Small 1	Mistral AI	71.3%
38	Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)	NVIDIA	70.7%
39	Gemini 2.5 Flash-Lite (Reasoning)	Google	70.3%
40	Magistral Medium 1	Mistral AI	70.0%
41	MoonshotAI: Kimi K2 0711	Moonshot AI	69.3%
42	Solar Pro 2 (Reasoning)	Upstage	69.0%
43	deepseek-ai/DeepSeek-R1-Distill-Qwen-32B	DeepSeek	68.7%
44	DeepSeek R1 (Jan '25)	DeepSeek	68.3%
45	Z.ai: GLM 4.5 Air	Z.ai	67.3%
46	deepseek-ai/DeepSeek-R1-Distill-Llama-70BHistorical	DeepSeek	67.0%
47	deepseek-ai/DeepSeek-R1-Distill-Qwen-14B	DeepSeek	66.7%
48	Solar Pro 2 (Preview) (Reasoning)	Upstage	66.3%
49	Qwen3 4B (Reasoning)	Alibaba	65.7%
50	DeepSeek R1 0528 Qwen3 8B	DeepSeek	65.0%

Access this data

JSON GET /api/v1

[
  {
    "rank": 1,
    "model": "GPT-5 (high)",
    "provider": "OpenAI",
    "score": "95.6666666666667"
  },
  {
    "rank": 2,
    "model": "Grok 4",
    "provider": "SpaceXAI",
    "score": "94.3333333333333"
  },
  {
    "rank": 3,
    "model": "OpenAI: o4 Mini High",
    "provider": "OpenAI",
    "score": "94.0"
  }
]

Curl

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

CLI brew install rightrank

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

Claude Code Skill Skill

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