Visual Understanding on HR-Bench 8K
86.6Avg@8 Exact MatchSenseNova-MARS-32B
Evaluation results
| Method | Links | |
|---|---|---|
| SenseNova-MARS-32BMethod Category=Agentic Model2025.12 | 86.6 | |
| Gemini-2.5-ProMethod Category=Direct Answer2025.12 | 85.4 | |
| Qwen3-VL-235B-A22B-InstructMethod Category=Agentic Model2025.12 | 82.4 | |
| Qwen3-VL-32B-InstructMethod Category=Agentic Model2025.12 | 81.6 | |
| Skywork-R1V4Method Category=Agentic Model2025.12 | 79.8 | |
| SenseNova-MARS-8BMethod Category=Agentic Model2025.12 | 78.4 | |
| Qwen3-VL-8B-InstructMethod Category=Direct Answer2025.12 | 74.6 | |
| DeepEyesV2Method Category=Agentic Model2025.12 | 73.8 | |
| Mini o3Method Category=Agentic Model2025.12 | 73.3 | |
| ThymeMethod Category=Agentic Model2025.12 | 72 | |
| DeepEyesMethod Category=Agentic Model2025.12 | 69.5 | |
| MonetMethod Category=Agentic Model2025.12 | 68 | |
| Pixel-ReasonerMethod Category=Agentic Model2025.12 | 66.1 | |
| Qwen2.5-VL-32B-InstructMethod Category=Direct Answer2025.12 | 63.6 | |
| Qwen2.5-VL-7B-InstructMethod Category=Direct Answer2025.12 | 62.1 | |
| LLaVA-onevisonMethod Category=Direct Answer2025.12 | 59.8 | |
| GPT-4oMethod Category=Direct Answer2025.12 | 59.6 |