Spatial Reasoning on BLINK J
80.67AccuracyInternVL3.5-38B
Evaluation Results
| Method | Links | |
|---|---|---|
| InternVL3.5-38B3-run average=true2026.06 | 80.67 | |
| GPT-53-run average=true2026.06 | 77.33 | |
| Visual-OPSD3-run average=true, Latency (s)=10.02026.06 | 77.33 | |
| GPT-4o3-run average=true2026.06 | 72.67 | |
| InternVL3.5-8B3-run average=true2026.06 | 71.33 | |
| Qwen3-VL-32B3-run average=true2026.06 | 70.66 | |
| Text-only SFT3-run average=true, Latency (s)=28.52026.06 | 68.67 | |
| Qwen3-VL-8B3-run average=true2026.06 | 68.66 | |
| Visual-OPSD-Noise3-run average=true, Latency (s)=14.92026.06 | 68.66 | |
| BAGEL-7B3-run average=true2026.06 | 67.33 | |
| ThinkMorph3-run average=true, Latency (s)=142.82026.06 | 66 | |
| Janus-Pro-7B3-run average=true2026.06 | 50.67 | |
| Chameleon-7B3-run average=true2026.06 | 0.67 |