--- license: apache-2.0 language: - en --- # MedM-VL-CT-3B-en ## Introduction A medical LVLM, trained on **English** data, accepts text and **a single 3D CT volume** as input, and text-based results as output, enabling tasks such as **report generation** and **medical VQA**. Here are the evaluation results on **M3D-Bench**:

Method	Report Generation				Medical VQA
Method	BLEU	ROUGE	METEOR	BERT-Score	Accuracy	BLEU	ROUGE	METEOR	BERT-Score
RadFM	12.23	16.49	11.57	87.93	19.79	16.39	26.13	21.33	88.72
M3D-LaMed	15.15	19.55	14.38	88.46	75.78	49.38	52.39	33.58	91.53
MedM-VL-CT-3B-en	49.81	52.45	49.27	90.38	80.12	56.56	59.96	39.75	92.85

## Quickstart Please refer to [MedM-VL](https://github.com/MSIIP/MedM-VL).