Benchmark
MATH-500
Math benchmarks measure how well models solve maths problems. Scores are percentages. Higher is better.
- Models measured
- 27
- Scores recorded
- 27
- Sources
- 1
- IQ weight
- 1.54
Score distribution
- 94.8%
- Mean
- 96.2%
- Median
- 3.5%
- Std dev
- 90.2%
- 10th percentile
- 98.1%
- 90th percentile
- 83.9%
- Min
This benchmark has a weight of 1.54 for chat models, measured over 27 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 | LongCat-Flash-Thinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 99.2% |
| 2 | Sarvam 105B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 98.6% |
| 3 | GLM 4.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 98.2% |
| 4 | GLM 4.5 Air· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 98.1% |
| 5 | Nemotron Nano 9b V2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 97.8% |
| 6 | Kimi K2 Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 97.4% |
| 7 | Llama 3.1 Nemotron Ultra 253B v1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 97% |
| 8 | Sarvam-30B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 97% |
| 9 | longcat-flash-lite· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 96.8% |
| 10 | MiniMax M1 80K· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 96.8% |
| 11 | Qwen3 14B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 96.8% |
| 12 | Llama 3.3 Nemotron Super 49B v1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 96.6% |
| 13 | LongCat Flash Chat· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 96.4% |
| 14 | Claude Sonnet 3.7· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 96.2% |
| 15 | MiniMax M1 40K· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 96% |
| 16 | DeepSeek R1 Zero· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 95.9% |
| 17 | Aihubmix Phi 4 Mini (reasoning)· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 94.6% |
| 18 | DeepSeek R1 Distill Llama 70b· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 94.5% |
| 19 | DeepSeek R1 Distill Qwen 32B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 94.3% |
| 20 | DeepSeek R1 Distill Qwen 14B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 93.9% |
| 21 | DeepSeek R1 Distill Qwen 7B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 92.8% |
| 22 | QwQ 32B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 90.6% |
| 23 | QwQ 32B Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 90.6% |
| 24 | DeepSeek V3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 90.2% |
| 25 | O1 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 90% |
| 26 | DeepSeek R1 Distill Llama 8B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 89.1% |
| 27 | DeepSeek R1 Distill Qwen 1.5B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 83.9% |
Data credits: Data by LLM Stats