OpenCompass Multimodal - MM-Vet: leaderboard
The "MM-Vet" subtask of OpenCompass Multimodal. OpenCompass multimodal leaderboard — Shanghai AI Lab's vision-language benchmark aggregate within the OpenCompass evaluation platform.
Metric: Score (%). Source: rank.opencompass.org.cn. 105 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | SenseNova-V6-5-Pro | 89.4 | |
| 2 | CongRong-v2.0 | 83.9 | |
| 3 | Gemini-2.5-Pro | 83.3 | |
| 4 | JT3.5 | 82.9 | |
| 5 | TeleMM-2.0-Thinking | 82.8 | |
| 6 | Step-1o | 82.8 | |
| 7 | InternVL3-8B | 82.8 | #620 |
| 8 | TeleMM-2.0 | 81.5 | |
| 9 | RBDashMM3-DPO-38B | 81.2 | |
| 10 | InternVL3-38B | 81.1 | #405 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-multimodal-mm-vet · How It Works · Data refreshed daily, snapshot 2026-10-20.