Multimodal Reasoning and Perception on MMStar (test)
63.9AccuracyGPT-4o
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4o2026.04 | 63.9 | |
| InternVL2.5-8B2026.04 | 63.2 | |
| Qwen2.5-VL-7B + Saliency-R1Training Stage=Saliency-R12026.04 | 62.6 | |
| Qwen2.5-VL-7BTraining Stage=Base2026.04 | 62.4 | |
| Claude-3.5 Sonnet2026.04 | 62.2 | |
| LLaVA-OneVision-7B2026.04 | 61.9 | |
| InternVL2-8B2026.04 | 61.5 | |
| Insight-V-8B2026.04 | 61.5 | |
| Vision-R1-LlamaV-CI-11B2026.04 | 61.4 | |
| Mulberry-7B2026.04 | 61.3 | |
| Qwen2.5-VL-7B + SFTTraining Stage=SFT2026.04 | 60.1 | |
| IXC-2.52026.04 | 59.9 | |
| Gemini-1.5-Pro2026.04 | 59.1 | |
| MiniCPM-V-2.6-8B2026.04 | 57.5 | |
| Qwen2.5-VL-3BTraining Stage=Base2026.04 | 56.1 | |
| Qwen2.5-VL-3B + Saliency-R1Training Stage=Saliency-R12026.04 | 56 | |
| Qwen2.5-VL-3B + SFTTraining Stage=SFT2026.04 | 55.7 | |
| LLaVA-CoT-11B2026.04 | 54.9 | |
| MiniCPM-Llama-V-2.5-8B2026.04 | 51.8 |