The AI tools index that doesn't waste your time.不浪费你时间的 AI 工具索引。
Math · weight 8%数学 · 权重 8%

AIME 2025

⚠ saturation watchlist饱和观察名单 official site官网 ↗

What it tests: Problems from the 2025 American Invitational Mathematics Examination — olympiad-level math that requires multi-step derivation, not pattern matching.

考什么:2025 年美国数学邀请赛真题——奥赛级数学题,需要多步推导而不是套题型。

Why it counts: The default math headline number vendors quote — but the best reasoning models now approach 95-99%, so we pair it with FrontierMath for headroom.

为什么算数:厂商最爱引用的数学成绩——但顶级推理模型已接近 95-99% 天花板,所以我们搭配 FrontierMath 保持区分度。

Maintainer维护方MAA(竞赛主办方),社区汇总
License许可题目版权 MAA,成绩公开引用
Reference point (prior-gen SOTA)参考点(上一代 SOTA)85
◈

Standings实时排名

53 models with sourced scores 个模型有溯源成绩
#Model模型Score成绩Index指数Dated日期Measured by测评方
1GPT-5.6 LunaOpenAI100100—3rd-party第三方source ↗×5
2Gemini 3.6 FlashGoogle100100—3rd-party第三方source ↗×3⚠±6
3Kimi K2.7 CodeMoonshot AI100100—3rd-party第三方source ↗
4LongCat 2.0Meituan100100—vendor-reported厂商自报source ↗×2
5Claude Fable 5Anthropic99.7100—3rd-party第三方source ↗×2⚠±8.7
6GLM 5.2Z.ai98.6992026-07-173rd-party第三方source ↗×6
7Claude Opus 4.8Anthropic98.398—3rd-party第三方source ↗×6
8Step 3.7 FlashStepFun97.397—3rd-party第三方source ↗
9GPT-5.6 SolOpenAI9797—3rd-party第三方source ↗×5⚠±3
10GPT-5.6 TerraOpenAI9797—3rd-party第三方source ↗
11Claude Opus 5Anthropic9797—3rd-party第三方source ↗×4⚠±27.1
12DeepSeek V4 Flash 0423DeepSeek9797—3rd-party第三方source ↗×3⚠±27
13DeepSeek V4 Pro 0423DeepSeek96.67⚡max97—3rd-party第三方source ↗×6
14Gemini 3.5 FlashGoogle96.6797—3rd-party第三方source ↗×3
15Kimi K3Moonshot AI96.296—3rd-party第三方source ↗×4
16Kimi K2 ThinkingMoonshot AI96.1⚡high962025-123rd-party第三方source ↗×6⚠±5.5
17GLM 5.1Z.ai95.395—3rd-party第三方source ↗×4
18gpt-oss-120bOpenAI93.493—3rd-party第三方source ↗×3⚠±4.5
19Gemini 3.1 Pro PreviewGoogle93.393—3rd-party第三方source ↗×6⚠±8.8
20MiniMax M3MiniMax93.393—3rd-party第三方source ↗×4⚠±25
21DeepSeek V4 Flash 0731DeepSeek93.1⚡max93—3rd-party第三方source ↗×3
22DeepSeek V3.2DeepSeek93.1⚡high93—3rd-party第三方source ↗×3⚠±13.6
23Seed-2.0-LiteByteDance Seed9393—3rd-party第三方source ↗×3
24Mistral Small 4Mistral AI9393—3rd-party第三方source ↗×4
25Grok 4.20xAI91.792—3rd-party第三方source ↗×3
26Grok 4.5xAI91.7⚡high92—3rd-party第三方source ↗×4
27Mercury 2Inception91.191—3rd-party第三方source ↗×3
28Command ACohere9090—3rd-party第三方source ↗
29Grok 4.3xAI8989—3rd-party第三方source ↗×3⚠±11
30Seed-2.0-MiniByteDance Seed8787—3rd-party第三方source ↗×2
31Mistral Medium 3.5Mistral AI86.386—3rd-party第三方source ↗×2⚠±46.3
32Qwen3.7 MaxQwen85.786—3rd-party第三方source ↗×5⚠±14.3
33Qwen3.7 PlusQwen85.786—3rd-party第三方source ↗×2⚠±18.4
34Mistral LargeMistral AI8585—3rd-party第三方source ↗×3⚠±71
35Solar Pro 3Upstage84.384—3rd-party第三方source ↗
36Qwen3.8 Max PrimeQwen81.582—3rd-party第三方source ↗×3
37GPT-5.5OpenAI81.2812026-05-053rd-party第三方source ↗×7⚠±18.8
38MiniMax M2.7MiniMax78.378—3rd-party第三方source ↗×2
39Hunyuan A13B InstructTencent76.877—3rd-party第三方source ↗×3
40Qwen3.7 FlashQwen67.367—3rd-party第三方source ↗
41Claude Sonnet 5Anthropic57.457—3rd-party第三方source ↗×3⚠±39.6
42Claude Haiku 4.5Anthropic3939—3rd-party第三方source ↗×4⚠±61
43MiMo-V2.5-ProXiaomi37.337—3rd-party第三方source ↗×2⚠±57.5
44Gemini 3.5 Flash LiteGoogle3636—3rd-party第三方source ↗×3⚠±60.67
45Muse Spark 1.1meta3030—vendor-reported厂商自报source ↗
46Gemini 3.1 Flash LiteGoogle3030—3rd-party第三方source ↗×5⚠±58.3
47Sonar ProPerplexity2929—3rd-party第三方source ↗×2⚠±48
48Llama 4 MaverickMeta2020—3rd-party第三方source ↗
49Nova Premier 1.0Amazon1717—3rd-party第三方source ↗×2
50Nova Pro 1.0Amazon77—3rd-party第三方source ↗×2⚠±90
51Command R+ (08-2024)Cohere6.77—3rd-party第三方source ↗
52Phi 4Microsoft6.77—3rd-party第三方source ↗×3⚠±56.2
53Llama 4 ScoutMeta6.77—3rd-party第三方source ↗

How scores become the Intelligence Score →成绩如何合成智能评分 →