Home
Information

AI Dataset Collection

Large-scale datasets and benchmarks for training, evaluating, and testing models to measure

Tools

Intelligent Document Recognition

Comprehensive Text Extraction and Document Processing Solutions for Users

AI Tutorial

visual-med-alpaca

Public

Visual Med-Alpaca is an open-source, multi-modal foundation model designed specifically for the biomedical domain, built on the LLaMa-7B.

Creat2023-04-11T23:31:14
Update2025-03-24T11:25:43
https://cambridgeltl.github.io/visual-med-alpaca/
391
Stars
0
Stars Increase