Perception Evaluation on MME Perception
1,755.5ScoreSAP
Evaluation Results
| Method | Links | |
|---|---|---|
| SAPBackbone=Qwen3-VL-8B, Strategy=SAP2026.02 | 1,755.5 | |
| Qwen3-VL-8B-InstructBackbone=Qwen3-VL-8B, Strategy=Instruct2026.02 | 1,744.7 | |
| VGAMLLM=Qwen2.5-VL-7B2025.11 | 1,719.7 | |
| TARACMLLM=Qwen2.5-VL-7B2025.11 | 1,713.1 | |
| VanillaMLLM=Qwen2.5-VL-7B2025.11 | 1,691.1 | |
| Qwen3-VL-8B-ThinkingBackbone=Qwen3-VL-8B, Strategy=LongCoT (Thinking)2026.02 | 1,663.2 | |
| OlympusLM=Phi-2-2.7B, Res.=3842024.12 | 1,520.7 | |
| LLaVA-1.5LM=Vicuna-7B, Res.=3362024.12 | 1,510.7 | |
| Mipha-3BLM=Phi-2-2.7B, Res.=3842024.12 | 1,488.9 | |
| Bunny-3BLM=Phi-2-2.7B, Res.=3842024.12 | 1,488.8 | |
| Qwen-VL-ChatLM=Qwen-7B, Res.=4482024.12 | 1,487.5 | |
| VGAMLLM=LLaVA-1.5-7B2025.11 | 1,465.7 | |
| TinyLLaVALM=Phi-2-2.7B, Res.=3842024.12 | 1,464.9 | |
| TARACMLLM=LLaVA-1.5-7B2025.11 | 1,462.1 | |
| VanillaMLLM=LLaVA-1.5-7B2025.11 | 1,456.5 | |
| mPLUG-Owl2LM=LLaMA-7B, Res.=4482024.12 | 1,450.2 | |
| MobileVLM-v2-3BLM=MobileLLaMA-2.7B, Res.=3362024.12 | 1,440.5 | |
| Imp-v1LM=Phi-2-2.7B, Res.=3842024.12 | 1,434 | |
| MoE-LLaVA-3.6BLM=Phi-2-2.7B, Res.=3842024.12 | 1,431.3 | |
| LLaVA-PhiLM=Phi-2-2.7B, Res.=3362024.12 | 1,335.1 | |
| MobileVLM-3BLM=MobileLLaMA-2.7B, Res.=3362024.12 | 1,288.9 |