Instructions to use prav-974/medical-qa-tinyllama with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use prav-974/medical-qa-tinyllama with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="prav-974/medical-qa-tinyllama")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("prav-974/medical-qa-tinyllama")
model = AutoModelForCausalLM.from_pretrained("prav-974/medical-qa-tinyllama")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use prav-974/medical-qa-tinyllama with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "prav-974/medical-qa-tinyllama"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "prav-974/medical-qa-tinyllama",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/prav-974/medical-qa-tinyllama

SGLang

How to use prav-974/medical-qa-tinyllama with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "prav-974/medical-qa-tinyllama" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "prav-974/medical-qa-tinyllama",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "prav-974/medical-qa-tinyllama" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "prav-974/medical-qa-tinyllama",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use prav-974/medical-qa-tinyllama with Docker Model Runner:
```
docker model run hf.co/prav-974/medical-qa-tinyllama
```

Model Card for Model ID

This model is a fine-tuned medical question-answering language model based on TinyLlama-1.1B-Chat, optimized using LoRA for efficient training on limited hardware.

Model Details

Model Description

This model is a medical Q&A instruction-tuned language model built by fine-tuning TinyLlama-1.1B-Chat using supervised fine-tuning (SFT) with LoRA.

It is designed to generate structured and informative responses to medical-related questions for educational and research purposes.

Developed by: Praveen
Funded by [optional]: Self / Academic Project
Shared by [optional]: Hugging Face Hub
Model type: Causal Language Model (Instruction-tuned)
Language(s) (NLP): English
License: Apache 2.0
Finetuned from model [optional]: TinyLlama/TinyLlama-1.1B-Chat-v1.0

Model Sources [optional]

Repository: https://huggingface.co//medical-qa-tinyllama
Paper [optional]: N/A
Demo [optional]: Included in inference code

Uses

Direct Use

Medical question answering
Educational assistance
Learning basic healthcare concepts

Downstream Use [optional]

Medical chatbots (non-clinical)
AI tutors for students
Research prototypes

Out-of-Scope Use

Clinical diagnosis
Emergency medical advice
Real-world healthcare decision-making
Any life-critical applications

Bias, Risks, and Limitations

May generate incorrect or outdated medical information
Limited dataset (~10K samples)
Not clinically validated
Can hallucinate plausible but incorrect answers
Lacks patient-specific reasoning

Recommendations

Use only for educational purposes
Always verify outputs with medical professionals
Do not deploy in high-risk environments
Apply safety filters if used in applications

How to Get Started with the Model

from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "pra-974/medical-qa-tinyllama"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

def ask(question):
    prompt = f"### Instruction:\n{question}\n\n### Response:\n"
    inputs = tokenizer(prompt, return_tensors="pt")

    outputs = model.generate(
        **inputs,
        max_new_tokens=256,
        temperature=0.7,
        top_p=0.9
    )

    return tokenizer.decode(outputs[0], skip_special_tokens=True)

print(ask("What are symptoms of diabetes?"))

Downloads last month: 39

Safetensors

Model size

1B params

Tensor type

F16