Metric Depth Estimation on NYU v2
93.5Delta 1 AccuracyVLM3-4b
Evaluation results
| Method | Links | |
|---|---|---|
| VLM3-4bModel Type=VLM Trained on Metric Depth Estimation2026.05 | 93.5 | |
| DepthLM-7bModel Type=VLM Trained on Metric Depth Estimation2026.05 | 91.5 | |
| DepthLM-3bModel Type=VLM Trained on Metric Depth Estimation2026.05 | 86.8 | |
| Seed1.5-VLModel Type=VLM Trained on Metric Depth Estimation2026.05 | 84.1 | |
| GPT-5Model Type=VLM2026.05 | 54 | |
| Gemini-2.5-ProModel Type=VLM2026.05 | 39.4 | |
| Qwen3-VL-32bModel Type=VLM2026.05 | 39.3 | |
| Qwen2.5-VL-72bModel Type=VLM2026.05 | 32.4 | |
| SpatialRGPT-8BModel Type=Spatial VLM2026.05 | 26.5 | |
| SpaceLLaVA-13BModel Type=Spatial VLM2026.05 | 17.8 | |
| Qwen3-VL-4bModel Type=VLM2026.05 | 15.8 |