Multimodal Evaluation on MME-RealWorld (Accuracy)
71.2AccuracyACE-Brain-0-8B
Evaluation Results
| Method | Links | |
|---|---|---|
| ACE-Brain-0-8BModel Type=Embodied Brain MLLM2026.03 | 71.2 | |
| Gemini2.5-proModel Type=Closed-source MLLM2026.03 | 67 | |
| Qwen3-VL-8B-Inst.Model Type=Open-source general-purpose MLLM2026.03 | 63.3 | |
| Qwen-VL-MaxModel Type=Closed-source MLLM2026.03 | 61.7 | |
| MiMo-Embodied-7BModel Type=Embodied Brain MLLM2026.03 | 60.3 | |
| VeBrain-7BModel Type=Embodied Brain MLLM2026.03 | 60.1 | |
| RoboBrain2.5-8BModel Type=Embodied Brain MLLM2026.03 | 60 | |
| RoboBrain2.0-7BModel Type=Embodied Brain MLLM2026.03 | 59.6 | |
| Qwen2.5-VL-7B-Inst.Model Type=Open-source general-purpose MLLM2026.03 | 58.6 | |
| GPT-4oModel Type=Closed-source MLLM2026.03 | 58 | |
| Pelican-VL-7BModel Type=Embodied Brain MLLM2026.03 | 57.9 | |
| MiMo-VL-7BModel Type=Open-source general-purpose MLLM2026.03 | 54.1 | |
| InternVL3-8BModel Type=Open-source general-purpose MLLM2026.03 | 52.1 | |
| InternVL3.5-8BModel Type=Open-source general-purpose MLLM2026.03 | 49.2 | |
| Vlaser-8BModel Type=Embodied Brain MLLM2026.03 | 41.6 | |
| LLaVA-OV-Pt + SFTBackbone=LLaVA-OV, Training stage=SFT2026.07 | 40.9 | |
| InternVL2.5-Pt + IRABackbone=InternVL2.5, Training stage=IRA2026.07 | 40.8 | |
| InternVL2-Pt + IRABackbone=InternVL2, Training stage=IRA2026.07 | 40.5 | |
| InternVL2.5-Pt + SFTBackbone=InternVL2.5, Training stage=SFT2026.07 | 40.4 | |
| LLaVA-OV-Pt + IRABackbone=LLaVA-OV, Training stage=IRA2026.07 | 40 | |
| InternVL2-Pt + SFTBackbone=InternVL2, Training stage=SFT2026.07 | 37.6 | |
| InternVL2-PtBackbone=InternVL2, Training stage=Pre-trained2026.07 | 31.2 | |
| InternVL2.5-PtBackbone=InternVL2.5, Training stage=Pre-trained2026.07 | 29.5 | |
| LLaVA-OV-PtBackbone=LLaVA-OV, Training stage=Pre-trained2026.07 | 28 |