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 |
|---|---|---|
| 51 | DeepSeek R1 Distill Qwen 1.5B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 52.7% |
| 52 | QwQ 32B Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 50% |
| 53 | GPT-4.1 mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 49.6% |
| 54 | GPT 4.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 48.1% |
| 55 | GPT OSS 20bofficial 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. | 42.1% |
| 56 | O1 Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 42% |
| 57 | DeepSeek V3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 39.2% |
| 58 | GPT-4.5 Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 36.7% |
| 59 | GPT-4.1 nano· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 29.4% |
| 60 | GPT 4o· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 13.1% |
Showing 51–60 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.