Benchmark
AIME Benchmark
What is AIME?
pass@k, majority vote and Python tool access each shift this by double digits. Benchmark Radar tracks 24 reported scores for AIME, cited by 19 source documents.
- Released
- 2024-02-01
- Evaluates
- math
- Openness
- unknown
- Importer
- Model reports
- Reported scores
- 24
Source provenance
- Original evidence https://maa.org/maa-invitational-competitions/
AIME links
AIME results and reported scores
Model reports
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Bespoke-Stratos-7B | Bespoke Labs | 20.0 | 2025-04-30 | evidence |
| Claude Opus 4 | Anthropic | 75.5 | 2025-05-22 | evidence |
| Claude Sonnet 4 | Anthropic | 70.5 | 2025-05-22 | evidence |
| DeepSeek-R1 | DeepSeek | 79.8 | 2025-01-22 | evidence |
| DeepSeek-R1-Distill-Llama-8B | DeepSeek | 43.3 | 2025-04-30 | evidence |
| DeepSeek-R1-Distill-Qwen-7B | DeepSeek | 53.3 | 2025-04-30 | evidence |
| DeepSeek-V3 | DeepSeek | 39.2 | 2024-12-27 | evidence |
| Gemma 4 (31B) | 89.2 | 2026-03-11 | evidence | |
| GLM-5 | Z.ai | 92.7 | 2026-02-11 | evidence |
| GLM-5.1 | Z.ai | 95.3 | 2026-04-03 | evidence |
| GLM-5.2 | Z.ai | 99.2 | 2026-06-16 | evidence |
| Llama-3.2-3B-Instruct | Meta | 6.7 | 2025-04-30 | evidence |
| o1-mini | OpenAI | 63.6 | 2025-04-30 | evidence |
| OpenThinker-7B | OpenThoughts | 31.3 | 2025-04-30 | evidence |
| Phi-4-Mini (base model, 3.8B) | Microsoft | 10.0 | 2025-04-30 | evidence |
| Phi-4-mini-reasoning (3.8B) | Microsoft | 57.5 | 2025-04-30 | evidence |
| Qwen3-235B-A22B (Thinking) | Qwen | 85.7 | 2025-05-14 | evidence |
| Qwen3.5-397B-A17B | Qwen | 91.3 | 2026-02-16 | evidence |
| Seed2.0 Lite | ByteDance | 92.3 | 2026-02-14 | evidence |
| Seed2.0 Lite | ByteDance | 86.5 | 2026-02-14 | evidence |
| Seed2.0 Mini | ByteDance | 84.6 | 2026-02-14 | evidence |
| Seed2.0 Mini | ByteDance | 80.8 | 2026-02-14 | evidence |
| Seed2.0 Pro | ByteDance | 98.3 | 2026-02-14 | evidence |
| Seed2.0 Pro | ByteDance | 94.2 | 2026-02-14 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.
Which sources cite AIME?
- GLM-5.2 Z.ai, 2026-06-16
- GLM-5.1 Z.ai, 2026-04-03
- Gemma 4 (31B) Google, 2026-03-11
- Qwen3.5-397B-A17B Qwen, 2026-02-16
- Seed2.0 (Pro / Lite / Mini) ByteDance, 2026-02-14
- GLM-5 Z.ai, 2026-02-11
- GLM-5 Z.ai, 2026-02-11
- Olmo 3 (7B and 32B) Ai2, 2025-11-20
- GPT-5 OpenAI, 2025-08-07
- Grok 4 xAI, 2025-07-09
- Gemini 2.5 Pro Google, 2025-06-17
- Claude Opus 4 and Sonnet 4 Anthropic, 2025-05-22
19 source documents cite this benchmark.