MMEB Leaderboard: leaderboard
Massive Multimodal Embedding Benchmark leaderboard evaluating image, video, and visual-document embedding models across classification, retrieval, QA, grounding, and document tasks.
Metric: Overall Score (%). Source: huggingface.co. 101 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Wianmu-5B-v0 | 60.45 | |
| 2 | WeMM-Embedding-9B | 59.33 | |
| 3 | Ovis-Omni-Embedding-3B | 58.46 | |
| 4 | Wianmu-3B-v0 | 58.16 | |
| 5 | WeMM-Embedding-4B | 57.97 | |
| 6 | Ovis-Omni-Embedding-v0.5-3B | 57.42 | |
| 7 | WeMM-Embedding-2B | 55.77 | |
| 8 | Tianmu-Emb-Uni | 53.27 | |
| 9 | Ovis-Omni-Embedding-v0.1-3B | 53.09 | |
| 10 | UEmbed-9B (hybrid) | 52.26 |
Interactive version: theaggregate.ai/benchmark?slug=mmeb-leaderboard · How It Works · Data refreshed daily, snapshot 2026-10-20.