metadata
datasets:
- BIOMEDICA/biomedica_webdataset_24M
language:
- en
base_model:
- HuggingFaceTB/SmolVLM-Instruct-500M
BMC-SmolVLM1 is a family of lightweight biomedical vision-language models (ranging from 256M to 2.2B parameters) based on SmolVLM. These models are designed for efficient multimodal understanding in the biomedical domain. Please ensure you are using a GPU runtime to run this notebook.
Colab Tutorial: 