yasserrmd/MedScholar-1.5B

MedScholar-1.5B is a 1.5 billion parameter instruction-aligned medical question-answering model developed by yasserrmd, based on the Qwen2.5-1.5B-Instruct architecture. Fine-tuned on 1 million examples from the MIRIAD-4.4M dataset, it is designed for efficient, in-context clinical knowledge exploration with a 131072 token context length. This model excels at providing concise answers to medical questions for research and educational purposes.