A benchmark for evaluating image-embedding capabilities across multimodal retrieval and representation tasks.
Browse the latest scores, model modes, release dates, and parameter sizes for MMEB-v2-Image.
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 | |||
|---|---|---|---|---|---|
![]() Qwen3-VL-Embedding-8B Standard Mode | 80.12 | 2026-01-08 | 8B | Free Commercial | |
![]() Seed1.6-Embedding-1215 Standard Mode | 77.99 | 2025-12-15 | Unknown | Closed | |
![]() Seed-1.6-Embedding-250615 Standard Mode | 77.78 | 2025-06-28 | Unknown | Closed | |
4 | ![]() Qwen3-VL-Embedding-2B Standard Mode | 74.96 | 2026-01-08 | 2B | Free Commercial |
5 | ![]() GME-Qwen2-VL-7B Standard Mode | 55.95 | 2024-12-24 | 7B | Free Commercial |
6 | ![]() GME-Qwen2-VL-2B Standard Mode | 51.89 | 2024-12-24 | 2B | Free Commercial |