Fine-grained visual understanding on HR-Bench 8K
74.9ScoreSwimBird
Evaluation results
| Method | Links | |
|---|---|---|
| SwimBirdBackbone=Qwen3-VL 8B2026.02 | 74.9 | |
| DeepEyesV2Model Category=Multimodal Agentic Models2026.02 | 73.8 | |
| ThymeModel Category=Multimodal Agentic Models2026.02 | 72 | |
| Qwen3-VL-8B-InstructModel Category=Textual Reasoning Models, Reproduced=true2026.02 | 71.3 | |
| DeepEyesModel Category=Multimodal Agentic Models2026.02 | 69.5 | |
| InternVL3-8BModel Category=Textual Reasoning Models2026.02 | 69.3 | |
| Qwen3-VL-8B-ThinkingModel Category=Textual Reasoning Models2026.02 | 68.1 | |
| MonetModel Category=Latent Visual Reasoning Models2026.02 | 68 | |
| SkiLaModel Category=Latent Visual Reasoning Models2026.02 | 66.5 | |
| LVRModel Category=Latent Visual Reasoning Models2026.02 | 66.1 | |
| Pixel ReasonerModel Category=Multimodal Agentic Models2026.02 | 66.1 | |
| Qwen2.5-VL-32B-InstructModel Category=Textual Reasoning Models2026.02 | 63.6 | |
| Qwen2.5-VL-7B-InstructModel Category=Textual Reasoning Models2026.02 | 62.1 | |
| GPT-5-miniModel Category=Textual Reasoning Models2026.02 | 60.9 | |
| LLaVA-OneVisonModel Category=Textual Reasoning Models2026.02 | 59.8 | |
| Vision-R1Model Category=Textual Reasoning Models2026.02 | 57 | |
| GPT-4oModel Category=Textual Reasoning Models2026.02 | 55.5 |