Medical Vision-Language Modeling on CT-FlowBench
58QA ScoreHulu-Med-7B
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Hulu-Med-7BTool-use=false2026.02 | 58 | 46 | 37 | 47 | |
| Gemini-3-Pro-PreviewTool-use=true2026.02 | 45 | 43 | 44 | 44 | |
| Claude-Sonnet-4.5Tool-use=true2026.02 | 44 | 44 | 43 | 43.67 | |
| CT-Flow-7BTool-use=true2026.02 | 43 | 52 | 38 | 44.33 | |
| CT-Flow-8BTool-use=true2026.02 | 42 | 40 | 47 | 43 | |
| M3D-RADTool-use=false2026.02 | 39 | 34 | 35 | 36 | |
| GPT-5.2Tool-use=true2026.02 | 35 | 40 | 37 | 37.33 | |
| Qwen3-VL-235B-A22B-InstructTool-use=true2026.02 | 30 | 36 | 36 | 34 | |
| Qwen3-VL-8B-InstructTool-use=true2026.02 | 30 | 26 | 20 | 25.33 | |
| GLM4.6-VTool-use=true2026.02 | 25 | 33 | 34 | 30.67 | |
| M3D-LaMed-Llama-2-7BTool-use=false2026.02 | 17 | 17 | 17 | 17 | |
| Qwen2.5-VL-7B-InstructTool-use=true2026.02 | 14 | 22 | 21 | 19 |