Multi-modal Spatial Intelligence on MMSI
38AccuracyGemini-2.5-Pro
Evaluation results
| Method | Links | |
|---|---|---|
| Gemini-2.5-ProModel Type=Closed-source MLLM2026.03 | 38 | |
| ACE-Brain-0-8BModel Type=Embodied Brain MLLM2026.03 | 32.2 | |
| MiMo-Embodied-7BModel Type=Embodied Brain MLLM, Evaluation Protocol=Our evaluation framework2026.03 | 31.7 | |
| Qwen3-VL-8B-Inst.Model Type=Open-source general-purpose MLLM, Evaluation Protocol=Our evaluation framework2026.03 | 31 | |
| GPT-4oModel Type=Closed-source MLLM2026.03 | 30.3 | |
| InternVL3.5-8BModel Type=Open-source general-purpose MLLM, Evaluation Protocol=Our evaluation framework2026.03 | 30.2 | |
| Qwen3-VL-2B-Inst.Model Type=Open-source general-purpose MLLM, Evaluation Protocol=Our evaluation framework2026.03 | 29.7 | |
| RoboBrain2.5-8BModel Type=Embodied Brain MLLM, Evaluation Protocol=Our evaluation framework2026.03 | 29.3 | |
| RoboBrain2.0-7BModel Type=Embodied Brain MLLM2026.03 | 27.9 | |
| VeBrain-7BModel Type=Embodied Brain MLLM, Evaluation Protocol=Our evaluation framework2026.03 | 27.3 | |
| Vlaser-8BModel Type=Embodied Brain MLLM2026.03 | 27.2 | |
| Qwen2.5-VL-7B-Inst.Model Type=Open-source general-purpose MLLM2026.03 | 26.8 | |
| Pelican-VL-7BModel Type=Embodied Brain MLLM, Evaluation Protocol=Our evaluation framework2026.03 | 26 | |
| InternVL3-8BModel Type=Open-source general-purpose MLLM2026.03 | 25.7 |