Benchmark
mu-MATH TPR
Math benchmarks measure how well models solve maths problems. Scores are percentages. Higher is better.
- Models measured
- 17
- Scores recorded
- 85
- Sources
- 1
- IQ weight
- 0.53
Score distribution
- 74.2%
- Mean
- 76.1%
- Median
- 12%
- Std dev
- 57.9%
- 10th percentile
- 90.5%
- 90th percentile
- 45.2%
- Min
This benchmark has a weight of 0.53 for chat models, measured over 17 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 | o1Gemini-1.5-Pro, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 95.3% |
| 2 | QwQ 32B PreviewGemini-1.5-Pro, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 95.3% |
| 3 | Gemini 2.0 Flash ExperimentalGemini-1.5-Pro, TPR, Qwen/Qwen2.5-72B-Instruct, thinking· U-MATH and mu-MATH | 93% |
| 4 | QwQ 32B PreviewQwen2.5-72B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 92.1% |
| 5 | Gemini 2.0 Flash ExperimentalGPT-4o, TPR, Qwen/Qwen2.5-72B-Instruct, thinking· U-MATH and mu-MATH | 91.7% |
| 6 | QwQ 32B Previewmu_math, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 91.3% |
| 7 | o1mu_math, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 90.6% |
| 8 | o1Llama-3.1-70B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 90.5% |
| 9 | QwQ 32B PreviewLlama-3.1-70B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 90.5% |
| 10 | Gemini 2.0 Flash Experimentalmu_math, TPR, Qwen/Qwen2.5-72B-Instruct, thinking· U-MATH and mu-MATH | 89.1% |
| 11 | o1GPT-4o, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 88.9% |
| 12 | O1 MiniGemini-1.5-Pro, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 88.4% |
| 13 | Gemini 2.0 Flash ExperimentalQwen2.5-72B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct, thinking· U-MATH and mu-MATH | 86.8% |
| 14 | o1Qwen2.5-72B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 86.8% |
| 15 | QwQ 32B PreviewGPT-4o, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 86.1% |
| 16 | DeepSeek R1Llama-3.1-70B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 85.7% |
| 17 | O1 MiniLlama-3.1-70B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 85.7% |
| 18 | Qwen2.5 Math 7B InstructLlama-3.1-70B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 85.6% |
| 19 | O1 Minimu_math, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 83.3% |
| 20 | Qwen2.5 Math 72B InstructGemini-1.5-Pro, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 82.5% |
| 21 | O1 MiniQwen2.5-72B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 81.6% |
| 22 | DeepSeek R1Gemini-1.5-Pro, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 81.4% |
| 23 | Qwen2.5 7B InstructGPT-4o, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 81.4% |
| 24 | Qwen2.5 Math 72B InstructLlama-3.1-70B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 81.4% |
| 25 | Qwen2.5 Math 72B InstructQwen2.5-72B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 81.3% |
| 26 | Gemini 1.5 ProGemini-1.5-Pro, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 81% |
| 27 | Gemini 2.0 Flash ExperimentalLlama-3.1-70B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct, thinking· U-MATH and mu-MATH | 81% |
| 28 | Qwen2.5 Math 72B Instructmu_math, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 80.9% |
| 29 | Mistral Large Instruct 2411Gemini-1.5-Pro, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 79.9% |
| 30 | Qwen2.5 72B InstructGemini-1.5-Pro, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 79.4% |
| 31 | Qwen2.5 7B InstructGemini-1.5-Pro, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 79.4% |
| 32 | Qwen2.5 7B Instructmu_math, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 78.7% |
| 33 | Qwen2.5 7B InstructLlama-3.1-70B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 78.4% |
| 34 | Qwen2.5 Math 72B InstructGPT-4o, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 77.9% |
| 35 | O1 MiniGPT-4o, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 77.8% |
| 36 | Qwen2.5 Math 7B InstructGemini-1.5-Pro, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 77.8% |
| 37 | Gemini 1.5 Promu_math, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 77.5% |
| 38 | Qwen2.5 72B Instructmu_math, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 77.1% |
| 39 | DeepSeek R1mu_math, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 76.8% |
| 40 | Qwen2.5 Math 7B Instructmu_math, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 76.6% |
| 41 | Gemini 1.5 ProGPT-4o, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 76.4% |
| 42 | Qwen2.5 72B InstructGPT-4o, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 76.4% |
| 43 | Qwen2.5 72B InstructQwen2.5-72B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 76.1% |
| 44 | Mistral Large Instruct 2411GPT-4o, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 75.7% |
| 45 | Mistral Large Instruct 2411mu_math, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 75.7% |
| 46 | Gemini 1.5 ProQwen2.5-72B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 75.5% |
| 47 | Qwen2.5 7B InstructQwen2.5-72B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 75.5% |
| 48 | Gemini 1.5 ProLlama-3.1-70B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 75.3% |
| 49 | Mistral Large Instruct 2411Llama-3.1-70B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 75.3% |
| 50 | Qwen2.5 72B InstructLlama-3.1-70B-Instruct, TPR, Qwen/Qwen2.5-72B-Instruct· U-MATH and mu-MATH | 75.3% |
Showing 1–50 of 85 results.
Data credits: Toloka, U-MATH and mu-MATH public leaderboard. MIT Space; generated solution outputs retain underlying model terms and are not ingested. (licence)