Holistic Multimodal Understanding on 14 Benchmarks Composite
65.2Average ScoreGenLIP
Evaluation Results
| Method | Links | |
|---|---|---|
| GenLIPArch=g/16, Data=8.0B2026.05 | 65.2 | |
| GenLIPArch=So/16, Data=8.0B2026.05 | 62.6 | |
| GenLIPArch=L/16, Data=8.0B2026.05 | 61.5 | |
| SigLIP2Arch=g/16, Data=40.0B2026.05 | 61.5 | |
| SigLIP2Arch=So/16, Data=40.0B2026.05 | 60.6 | |
| OVision2Arch=L/16, Data=12.8B2026.05 | 58.7 | |
| SigLIP2Arch=L/16, Data=40.0B2026.05 | 58.7 | |
| SigLIPArch=L/16, Data=40.0B2026.05 | 56.9 | |
| AIMv2Arch=L/14, Data=12.0B2026.05 | 55.7 | |
| CLIPArch=L/14, Data=12.8B2026.05 | 53.1 |