MathVision is an AI benchmark used to evaluate model capabilities. Review its overview, metrics, official resources, and model leaderboard results on DataLearnerAI.
Browse the latest scores, model modes, release dates, and parameter sizes for MathVision.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
| Rank | Model | License | |||
|---|---|---|---|---|---|
 Kimi K3 Thinking Level · MaxTools | 97.80 | 2026-07-16 | 2800B | Conditional | |
 Qwen3.8-Flash-Next Thinking Level · Extra HighTools | 95.70 | 2026-08-26 | 125B | Conditional | |
Qwen3.8-27B Thinking EnabledTools | 94.60 | 2026-08-14 | 27B | Free Commercial | |
4 |  Seed2.1 Pro Thinking EnabledTools | 94.50 | 2026-06-23 | Unknown | Closed |
5 | Kimi K3 Thinking Level · Max | 94.30 | 2026-07-16 | 2800B | Conditional |
6 | Seed2.1 Pro Thinking Enabled | 92.60 | 2026-06-23 | Unknown | Closed |
7 | Qwen3.8-Flash-Next Thinking Level · Extra High | 90.60 | 2026-08-26 | 125B | Conditional |
8 | Qwen3.8-27B Thinking Enabled | 90.00 | 2026-08-14 | 27B | Free Commercial |
9 |  Gemma 4 31B Thinking Enabled | 85.60 | 2026-04-02 | 30.7B | Free Commercial |
10 | Gemma 4 26B A4B Thinking Enabled | 82.40 | 2026-04-02 | 25.2B | Free Commercial |
11 | Gemma 4 E4B Thinking Enabled | 59.50 | 2026-04-02 | 8B | Free Commercial |
12 | Gemma 4 E2B Thinking Enabled | 52.40 | 2026-04-02 | 5.1B | Free Commercial |
