Benchmark
AIME 2024
Math benchmarks measure how well models solve maths problems. Scores are percentages. Higher is better.
- Models measured
- 50
- Scores recorded
- 60
- Sources
- 2
- IQ weight
- 0.61
Score distribution
- 75.7%
- Mean
- 80.4%
- Median
- 18.4%
- Std dev
- 48.1%
- 10th percentile
- 93.3%
- 90th percentile
- 13.1%
- Min
This benchmark has a weight of 0.61 for chat models, measured over 48 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | GPT OSS 120Bofficial model card Table 3 · reasoning high · with tools; Table 3 does not name the exact tool · pass@1· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 96.6% |
| 2 | GPT OSS 20bofficial model card Table 3 · reasoning high · with tools; Table 3 does not name the exact tool · pass@1· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 96% |
| 3 | GPT OSS 120Bofficial model card Table 3 · reasoning high · no tools · pass@1· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 95.8% |
| 4 | Grok 3 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 95.8% |
| 5 | o4 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 93.4% |
| 6 | Grok 3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 93.3% |
| 7 | LongCat-Flash-Thinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 93.3% |
| 8 | GPT OSS 20bofficial model card Table 3 · reasoning high · no tools · pass@1· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 92.1% |
| 9 | Gemini 2.5 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 92% |
| 10 | OpenAI o3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 91.6% |
| 11 | DeepSeek R1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 91.4% |
| 12 | GLM 4.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 91% |
| 13 | Ministral 14Blatest· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 89.8% |
| 14 | GLM 4.5 Air· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 89.4% |
| 15 | Gemini 2.5 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 88% |
| 16 | GPT OSS 120Bofficial model card Table 3 · reasoning medium · with tools; Table 3 does not name the exact tool · pass@1· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 87.9% |
| 17 | o3 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 87.3% |
| 18 | DeepSeek R1 Distill Llama 70b· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 86.7% |
| 19 | DeepSeek R1 Zero· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 86.7% |
| 20 | GPT OSS 20bofficial model card Table 3 · reasoning medium · with tools; Table 3 does not name the exact tool · pass@1· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 86% |
| 21 | MiniMax M1 80K· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 86% |
| 22 | Ministral 8Blatest· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 86% |
| 23 | O1 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 86% |
| 24 | Qwen 3 235B A22B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 85.7% |
| 25 | DeepSeek R1 Distill Qwen 32B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 83.3% |
| 26 | DeepSeek R1 Distill Qwen 7B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 83.3% |
| 27 | MiniMax M1 40K· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 83.3% |
| 28 | Qwen3 32B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 81.4% |
| 29 | Phi 4 Reasoning Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 81.3% |
| 30 | GPT OSS 120Bofficial model card Table 3 · reasoning medium · no tools · pass@1· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80.4% |
| 31 | Qwen3 30B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 80.4% |
| 32 | Claude Sonnet 3.7· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 80% |
| 33 | DeepSeek R1 Distill Llama 8B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 80% |
| 34 | DeepSeek R1 Distill Qwen 14B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 80% |
| 35 | GPT OSS 20bofficial model card Table 3 · reasoning medium · no tools · pass@1· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80% |
| 36 | QwQ 32B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 79.5% |
| 37 | Qwen3 14B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 79.3% |
| 38 | Ministral 3Blatest· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 77.5% |
| 39 | GPT OSS 120Bofficial model card Table 3 · reasoning low · with tools; Table 3 does not name the exact tool · pass@1· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 75.4% |
| 40 | Phi-4-reasoning· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 75.3% |
| 41 | o1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 74.3% |
| 42 | Magistral Medium· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 73.6% |
| 43 | Gemini 2.0 Flashthinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 73.3% |
| 44 | longcat-flash-lite· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 72.2% |
| 45 | Kimi K2 0905· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 72% |
| 46 | Magistral Small 25092506· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 70.7% |
| 47 | Kimi K2 Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 69.6% |
| 48 | DeepSeek V3.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 66.3% |
| 49 | GPT OSS 20bofficial model card Table 3 · reasoning low · with tools; Table 3 does not name the exact tool · pass@1· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 61.2% |
| 50 | GPT OSS 120Bofficial model card Table 3 · reasoning low · no tools · pass@1· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 56.3% |
Showing 1–50 of 60 results.
Data credits: Data by LLM Stats · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.