We-Math — leaderboard
Metric: Avg Strict (%). Source: we-math.github.io. 33 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 78 |
| 2 | Doubao-1.5-Pro | 65.7 |
| 3 | GPT-4.1 (2025-04-14) | 55.5 |
| 4 | GPT-4o | 50.6 |
| 5 | InternVL3-78B | 46.1 |
| 6 | InternVL3-8B | 39.5 |
| 7 | Qwen 2 VL 72B | 36.57 |
| 8 | InternVL2-8B | 26.57 |
| 9 | Gemini 1.5 Pro | 26.38 |
| 10 | Qwen 2 VL 7B | 25.62 |
Interactive version: theaggregate.ai/benchmark?slug=we-math · How the rankings work · Data refreshed daily, snapshot 2026-07-22.