Fine-grained Visual Search on V*
91.1Overall ScoreQwen2.5-VL + V-ABS
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Qwen2.5-VL + V-ABSBackbone=Qwen2.5-VL, Plugin=V-ABS2026.05 | 91.1 | 93 | 88.2 | |
| Qwen3-VL + V-ABSBackbone=Qwen3-VL, Plugin=V-ABS2026.05 | 90.5 | 91.2 | 89.3 | |
| Qwen2.5-VL + ZoomEyeBackbone=Qwen2.5-VL, Plugin=ZoomEye2026.05 | 89 | 88.7 | 89.4 | |
| Intern-VL3 + V-ABSBackbone=Intern-VL3, Plugin=V-ABS2026.05 | 85.9 | 90.4 | 79 | |
| Qwen2.5-VL + Pixel-ReasonerBackbone=Qwen2.5-VL, Plugin=Pixel-Reasoner2026.05 | 84.3 | — | — | |
| Qwen2.5-VL + ThymeBackbone=Qwen2.5-VL, Plugin=Thyme2026.05 | 82.2 | 83.5 | 80.3 | |
| Qwen2.5-VL + DeepEyes v2Backbone=Qwen2.5-VL, Plugin=DeepEyes v22026.05 | 81.8 | — | — | |
| GPT-4o + V-ABSBackbone=GPT-4o, Plugin=V-ABS2026.05 | 77 | 83.5 | 67.1 | |
| Qwen2.5-VL-7BModel=Qwen2.5-VL, Parameters=7B2026.05 | 75.9 | 76.5 | 75 | |
| Qwen3-VLModel=Qwen3-VL2026.05 | 75.9 | 73 | 80.3 | |
| Intern-VL3Model=Intern-VL32026.05 | 72.8 | 73 | 72.4 | |
| GPT-4oModel=GPT-4o2026.05 | 72.3 | 67 | 80.3 | |
| LLaVA-v1.6-13BModel=LLaVA-v1.6, Parameters=13B2026.05 | 61.8 | 60 | 64.5 | |
| LLaVA-HR-X-7BModel=LLaVA-HR-X, Parameters=7B2026.05 | 56.5 | 51.3 | 64.5 |