Spatial Reasoning on BLINK-R
87.1AccuracyQwen3-VL-8B
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen3-VL-8Bparameters=8B, zero-shot=true2025.12 | 87.1 | |
| RoboBrain2.0-7Bparameters=7B, zero-shot=true2025.12 | 83.87 | |
| GenieReasonerparameters=3B, zero-shot=true2025.12 | 83.87 | |
| InternVL-3.5-8Bparameters=8B, zero-shot=true2025.12 | 79.84 | |
| Cosmos-Reason1-7Bparameters=7B, zero-shot=true2025.12 | 79.84 | |
| Qwen2.5-VL-7Bparameters=7B, zero-shot=true2025.12 | 79.03 | |
| GPT-4o-minizero-shot=true2025.12 | 79.03 | |
| Gemini-2.5-prozero-shot=true2025.12 | 79.03 | |
| Claude-Sonnet-4zero-shot=true2025.12 | 72.58 | |
| Qwen2.5-VL-3Bparameters=3B, zero-shot=true2025.12 | 66.94 |