Visual Understanding on MME RealWorld
72.7Pass@1 Exact MatchSenseNova-MARS-32B
Evaluation Results
| Method | Links | |
|---|---|---|
| SenseNova-MARS-32BMethod Category=Agentic Model2025.12 | 72.7 | |
| Skywork-R1V4Method Category=Agentic Model2025.12 | 71.4 | |
| SenseNova-MARS-8BMethod Category=Agentic Model2025.12 | 67.9 | |
| Mini o3Method Category=Agentic Model2025.12 | 65.5 | |
| DeepEyesV2Method Category=Agentic Model2025.12 | 64.9 | |
| ThymeMethod Category=Agentic Model2025.12 | 64.8 | |
| Pixel-ReasonerMethod Category=Agentic Model2025.12 | 64.4 | |
| DeepEyesMethod Category=Agentic Model2025.12 | 64.1 | |
| GPT-4oMethod Category=Direct Answer2025.12 | 62.8 | |
| Qwen3-VL-8B-InstructMethod Category=Direct Answer2025.12 | 61.9 | |
| Qwen2.5-VL-32B-InstructMethod Category=Direct Answer2025.12 | 59.1 | |
| LLaVA-onevisonMethod Category=Direct Answer2025.12 | 57.4 | |
| Qwen2.5-VL-7B-InstructMethod Category=Direct Answer2025.12 | 56.8 |