This model is a medical-domain adapted conversational language model built using Parameter-Efficient Fine-Tuning (LoRA) on top of TinyLlama-1.1B-Chat. The model is trained to generate concise final medical diagnoses (≤5 words) from open-ended, verifiable medical questions.
This model is intended for research and experimentation in medical-domain adaptation of large language models, specifically for generating short diagnostic-style answers from structured medical questions.
The model can be integrated into research prototypes, educational tools, or experimental medical QA systems requiring concise diagnostic outputs.
It is a research prototype and not a certified medical tool.
Outputs should always be verified by qualified medical professionals. This model should be used strictly for research and educational purposes.
1from transformers import AutoTokenizer, AutoModelForCausalLM
2from peft import PeftModel
3
4base_model = AutoModelForCausalLM.from_pretrained(
5 "TinyLlama/TinyLlama-1.1B-Chat-v1.0"
6)
7
8model = PeftModel.from_pretrained(
9 base_model,
10 "mostafaalimohmed/gpt-oss-medical-qa"
11)
12
13tokenizer = AutoTokenizer.from_pretrained(
14 "mostafaalimohmed/gpt-oss-medical-qa"
15)