# AIME 2025 leaderboard

Mathematics benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelchoice.dev/leaderboards/benchmarks/math-aime2025

American Invitational Mathematics Examination 2025. Every model the catalog carries a published AIME 2025 value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Mathematics |
| Measure | Integer answers 000-999 |
| Tasks | 15 problems |
| Difficulty | High school olympiad level |
| Published by | [American Invitational Mathematics Examination](https://www.maa.org/math-competitions/aime) |

The most recent AIME examination, featuring 15 challenging mathematics problems testing olympiad-level mathematical reasoning with integer answers from 000-999.

## AIME 2025 ranking

55 models with a published AIME 2025 value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.

| Rank | Model | Provider | Integer answers 000-999 | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [GPT-5.2](https://modelchoice.dev/models/gpt-5-2) | OpenAI | 100 | Observed 2026-11-21 · source benchlm:models |
| 2 | [Ripple-65B Preview](https://modelchoice.dev/models/ripple-65b-preview) | Ricursive Intelligence | 98.3 | Observed 2026-11-21 · source benchlm:models |
| 3 | [Step 3.5 Flash](https://modelchoice.dev/models/step-3-5-flash) | StepFun | 97.3 | Observed 2026-11-21 · source benchlm:models |
| 4 | [MAI-Thinking-1](https://modelchoice.dev/models/mai-thinking-1) | Microsoft | 97 | Observed 2026-11-21 · source benchlm:models |
| 5 | [OrcaCyber-Nano-1.0](https://modelchoice.dev/models/orcacyber-nano-1-0) | OrcaRouter | 96.23 | Observed 2026-11-21 · source benchlm:models |
| 6 | [Kimi K2.5](https://modelchoice.dev/models/kimi-k2-5) | Moonshot AI | 96.1 | Observed 2026-11-21 · source benchlm:models |
| 6 | [Kimi K2.5 (Reasoning)](https://modelchoice.dev/models/kimi-k2-5-reasoning) | Moonshot AI | 96.1 | Last good 2026-11-18 · source benchlm:models |
| 8 | [GLM-4.7](https://modelchoice.dev/models/glm-4-7) | Z.AI | 95.7 | Observed 2026-11-21 · source benchlm:models |
| 9 | [Ternary Bonsai 2 27B](https://modelchoice.dev/models/ternary-bonsai-2-27b) | Prism ML | 95 | Observed 2026-11-21 · source benchlm:models |
| 10 | [MiMo-V2-Flash](https://modelchoice.dev/models/mimo-v2-flash) | Xiaomi | 94.1 | Observed 2026-11-21 · source benchlm:models |
| 11 | [DeepSeek V3.2 (Thinking)](https://modelchoice.dev/models/deepseek-v3-2-thinking) | DeepSeek | 93.1 | Last good 2026-11-18 · source benchlm:models |
| 12 | [o4-mini](https://modelchoice.dev/models/o4-mini) | OpenAI | 92.7 | Observed 2026-11-21 · source benchlm:models |
| 12 | [o4-mini (high)](https://modelchoice.dev/models/o4-mini-high) | OpenAI | 92.7 | Last good 2026-11-18 · source benchlm:models |
| 14 | [GPT-OSS 120B](https://modelchoice.dev/models/gpt-oss-120b) | OpenAI | 92.5 | Observed 2026-11-21 · source benchlm:models |
| 15 | [Qwen3 235B 2507 (Reasoning)](https://modelchoice.dev/models/qwen3-235b-2507-reasoning) | Alibaba | 92.3 | Last good 2026-11-18 · source benchlm:models |
| 16 | [GPT-OSS 20B](https://modelchoice.dev/models/gpt-oss-20b) | OpenAI | 91.7 | Observed 2026-11-21 · source benchlm:models |
| 17 | [GLM-4.7-Flash](https://modelchoice.dev/models/glm-4-7-flash) | Z.AI | 91.6 | Observed 2026-11-21 · source benchlm:models |
| 18 | [Nemotron 3 Super 100B](https://modelchoice.dev/models/nemotron-3-super-100b) | NVIDIA | 90.21 | Observed 2026-11-21 · source benchlm:models |
| 19 | [Granite 4.2 30B](https://modelchoice.dev/models/granite-4-2-30b) | IBM | 89.17 | Observed 2026-11-21 · source benchlm:models |
| 20 | [Nemotron 3 Nano 30B](https://modelchoice.dev/models/nemotron-3-nano-30b) | NVIDIA | 89.1 | Observed 2026-11-21 · source benchlm:models |
| 21 | [DeepSeek V3.1 (Reasoning)](https://modelchoice.dev/models/deepseek-v3-1-reasoning) | DeepSeek | 88.4 | Last good 2026-11-18 · source benchlm:models |
| 22 | [Sarvam 105B](https://modelchoice.dev/models/sarvam-105b) | Sarvam | 88.3 | Observed 2026-11-21 · source benchlm:models |
| 23 | [DeepSeek-R1-0528](https://modelchoice.dev/models/deepseek-r1-0528) | DeepSeek | 87.5 | Observed 2026-11-21 · source benchlm:models |
| 24 | [Claude Sonnet 4.5](https://modelchoice.dev/models/claude-sonnet-4-5) | Anthropic | 87 | Observed 2026-11-21 · source benchlm:models |
| 25 | [Granite 4.2 8B](https://modelchoice.dev/models/granite-4-2-8b) | IBM | 86.67 | Observed 2026-11-21 · source benchlm:models |
| 26 | [MiniCPM5-2B](https://modelchoice.dev/models/minicpm5-2b) | OpenBMB | 86.5 | Observed 2026-11-21 · source benchlm:models |
| 27 | [Exaone 4.0 32B](https://modelchoice.dev/models/exaone-4-0-32b) | LG AI Research | 85.3 | Observed 2026-11-21 · source benchlm:models |
| 28 | [Ministral 3 14B (Reasoning)](https://modelchoice.dev/models/ministral-3-14b-reasoning) | Mistral | 85 | Last good 2026-11-18 · source benchlm:models |
| 29 | [Nemotron 3 Nano Omni 30B A3B](https://modelchoice.dev/models/nemotron-3-nano-omni-30b-a3b) | NVIDIA | 82.1 | Observed 2026-11-21 · source benchlm:models |
| 30 | [Qwen3 235B-A22B (Reasoning)](https://modelchoice.dev/models/qwen3-235b-a22b-reasoning) | Alibaba | 81.5 | Observed 2026-11-21 · source benchlm:models |
| 31 | [Granite 4.2 3B](https://modelchoice.dev/models/granite-4-2-3b) | IBM | 78.33 | Observed 2026-11-21 · source benchlm:models |
| 32 | [Phi-4-reasoning-plus](https://modelchoice.dev/models/phi-4-reasoning-plus) | Microsoft | 78 | Observed 2026-11-21 · source benchlm:models |
| 33 | [MiniMax M1 80k](https://modelchoice.dev/models/minimax-m1-80k) | MiniMax | 76.9 | Observed 2026-11-21 · source benchlm:models |
| 34 | [DeepSeek-R1-0528-Qwen3-8B](https://modelchoice.dev/models/deepseek-r1-0528-qwen3-8b) | DeepSeek | 76.3 | Observed 2026-11-21 · source benchlm:models |
| 35 | [Claude 4 Opus Thinking](https://modelchoice.dev/models/claude-4-opus-thinking) | Anthropic | 75.5 | Observed 2026-11-21 · source benchlm:models |
| 36 | [Qwen3 32B (Reasoning)](https://modelchoice.dev/models/qwen3-32b-reasoning) | Alibaba | 72.9 | Observed 2026-11-21 · source benchlm:models |
| 37 | [Nemotron Ultra 253B](https://modelchoice.dev/models/nemotron-ultra-253b) | NVIDIA | 72.5 | Observed 2026-11-21 · source benchlm:models |
| 38 | [Qwen3 235B 2507](https://modelchoice.dev/models/qwen3-235b-2507) | Alibaba | 70.3 | Observed 2026-11-21 · source benchlm:models |
| 39 | [QwQ-32B](https://modelchoice.dev/models/qwq-32b) | Alibaba | 69.5 | Observed 2026-11-21 · source benchlm:models |
| 40 | [Magistral Medium](https://modelchoice.dev/models/magistral-medium) | Mistral | 64.9 | Observed 2026-11-21 · source benchlm:models |
| 41 | [Phi-4-reasoning](https://modelchoice.dev/models/phi-4-reasoning) | Microsoft | 62.9 | Observed 2026-11-21 · source benchlm:models |
| 42 | [Magistral Small](https://modelchoice.dev/models/magistral-small) | Mistral | 62.8 | Observed 2026-11-21 · source benchlm:models |
| 43 | [LFM2.5-2.6B](https://modelchoice.dev/models/lfm2-5-2-6b) | LiquidAI | 51.87 | Observed 2026-11-21 · source benchlm:models |
| 44 | [DeepSeek V3.1](https://modelchoice.dev/models/deepseek-v3-1) | DeepSeek | 49.8 | Observed 2026-11-21 · source benchlm:models |
| 44 | [Gemini 2.5 Flash-Lite](https://modelchoice.dev/models/gemini-2-5-flash-lite) | Google | 49.8 | Observed 2026-11-21 · source benchlm:models |
| 46 | [Kimi K2](https://modelchoice.dev/models/kimi-k2) | Moonshot AI | 49.5 | Observed 2026-11-21 · source benchlm:models |
| 47 | [LFM2.5-8B-A1B](https://modelchoice.dev/models/lfm2-5-8b-a1b) | LiquidAI | 42.53 | Observed 2026-11-21 · source benchlm:models |
| 48 | [MiniCPM5-1B](https://modelchoice.dev/models/minicpm5-1b) | OpenBMB | 40.42 | Observed 2026-11-21 · source benchlm:models |
| 49 | [Claude 4 Opus](https://modelchoice.dev/models/claude-4-opus) | Anthropic | 33.9 | Observed 2026-11-21 · source benchlm:models |
| 50 | [Claude 4 Sonnet](https://modelchoice.dev/models/claude-4-sonnet) | Anthropic | 33.1 | Observed 2026-11-21 · source benchlm:models |
| 51 | [Gemini 2.0 Flash](https://modelchoice.dev/models/gemini-2-0-flash) | Google | 29.7 | Observed 2026-11-21 · source benchlm:models |
| 52 | [Qwen3 235B-A22B](https://modelchoice.dev/models/qwen3-235b-a22b) | Alibaba | 24.7 | Observed 2026-11-21 · source benchlm:models |
| 53 | [Gemini 2.0 Flash-Lite](https://modelchoice.dev/models/gemini-2-0-flash-lite) | Google | 23.8 | Observed 2026-11-21 · source benchlm:models |
| 54 | [Qwen3 32B](https://modelchoice.dev/models/qwen3-32b) | Alibaba | 20.2 | Observed 2026-11-21 · source benchlm:models |
| 55 | [Gemini 1.5 Flash](https://modelchoice.dev/models/gemini-1-5-flash) | Google | 14.7 | Observed 2026-11-21 · source benchlm:models |

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
