U-MATH - Sequences & Series — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 28 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O3 Mini | 93.51 |
| 2 | DeepSeek R1 | 92.86 |
| 3 | O1 | 92.86 |
| 4 | Gemini 2.0 Flash (01-21) (Thinking) | 88.31 |
| 5 | QwQ 32B-Preview | 84.42 |
| 6 | O1 Mini | 76.62 |
| 7 | DeepSeek V3 | 72.08 |
| 8 | Qwen2.5-Math-72B-Instruct | 68.83 |
| 9 | QVQ-72B-Preview | 66.23 |
| 10 | Gemini 1.5 Pro | 62.99 |
| 11 | Gemini 1.5 Flash | 60.39 |
| 12 | Athene-V2-Chat | 57.79 |
| 13 | Qwen 2.5 32B Instruct | 56.49 |
| 14 | Mistral Large 2 (Nov) Instruct (2411) | 55.19 |
| 15 | Qwen 2.5 72B Instruct | 54.55 |
Interactive version: aibenchmarks.dev/benchmark?slug=u-math-sequences-series · How the rankings work · Data refreshed daily, snapshot 2026-07-08.