Visual Search on V-star
84.8AccuracyQwen3-VL-32B
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3-VL-32B3-run average=true, External visual-search tools usage=false2026.06 | 84.8 | |
| Qwen3-VL-8B3-run average=true, External visual-search tools usage=false2026.06 | 84.29 | |
| Penguin-VLParameters=2B2026.03 | 83.8 | |
| InternVL3.5-38B3-run average=true, External visual-search tools usage=false2026.06 | 76.96 | |
| Qwen3-VLParameters=2B2026.03 | 74.9 | |
| GPT-53-run average=true, External visual-search tools usage=false2026.06 | 71.73 | |
| InternVL3.5Parameters=2B2026.03 | 69.1 | |
| InternVL3.5-8B3-run average=true, External visual-search tools usage=false2026.06 | 68.59 | |
| ThinkMorph3-run average=true, Latency (s)=142.82026.06 | 67.01 | |
| Visual-OPSD3-run average=true, Latency (s)=10.02026.06 | 64.92 | |
| GPT-4o3-run average=true, External visual-search tools usage=false2026.06 | 61.78 | |
| Visual-OPSD-Noise3-run average=true, Latency (s)=14.92026.06 | 61.7 | |
| Text-only SFT3-run average=true, Latency (s)=28.52026.06 | 56.02 | |
| BAGEL-7B3-run average=true2026.06 | 55.49 | |
| SmolVLM2Parameters=2.2B2026.03 | 51.8 | |
| Gemma3n E2B-itParameters=E2B-it, Inference template=Penguin's template2026.03 | 46 | |
| Janus-Pro-7B3-run average=true2026.06 | 38.22 | |
| Chameleon-7B3-run average=true2026.06 | 28.27 |