Semantic Reasoning on DriveBench
53.18ScoreXEmbodied
Evaluation Results
| Method | Links | |
|---|---|---|
| XEmbodiedversion=Best2026.04 | 53.18 | |
| Qwen2.5-VL-32BCategory=Open-Source Models2026.04 | 53.06 | |
| Mimo-EmbodiedCategory=Embodied Models2026.04 | 52.95 | |
| PR1-Grounding-2BCategory=Spatial Models2026.04 | 52.03 | |
| Qwen3-VL-32BCategory=Open-Source Models2026.04 | 51.7 | |
| PR1-OCR-2BCategory=Spatial Models2026.04 | 50.73 | |
| UniVG-R1-7BCategory=Spatial Models2026.04 | 48.75 | |
| GPT-4oCategory=Proprietary Models2026.04 | 47.97 | |
| Qwen3-VL-A3B-30BCategory=Open-Source Models2026.04 | 46.22 | |
| DriveMMCategory=Embodied Models2026.04 | 44.5 | |
| Qwen2.5-VL-7BCategory=Open-Source Models2026.04 | 43.29 | |
| PR1-Detection-3BCategory=Spatial Models2026.04 | 41.35 | |
| PR1-Counting-2BCategory=Spatial Models2026.04 | 39.72 | |
| Cosmos-R1Category=Embodied Models2026.04 | 35.8 |