Instructions to use hackint0sh/phi-3-clinical with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hackint0sh/phi-3-clinical with Transformers:

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

pipe = pipeline("text-generation", model="hackint0sh/phi-3-clinical", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("hackint0sh/phi-3-clinical", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("hackint0sh/phi-3-clinical", trust_remote_code=True)
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 hackint0sh/phi-3-clinical with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "hackint0sh/phi-3-clinical"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hackint0sh/phi-3-clinical",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/hackint0sh/phi-3-clinical

SGLang

How to use hackint0sh/phi-3-clinical 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 "hackint0sh/phi-3-clinical" \
    --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": "hackint0sh/phi-3-clinical",
		"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 "hackint0sh/phi-3-clinical" \
        --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": "hackint0sh/phi-3-clinical",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use hackint0sh/phi-3-clinical with Docker Model Runner:
```
docker model run hf.co/hackint0sh/phi-3-clinical
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

🤖 Phi-3-Clinical

Welcome to the repository for Phi-3-Clinical, a fine-tuned model designed to empower medical researchers and developers in the Bio-Pharma domain. This model has been meticulously trained on clinical trial datasets from the U.S. government to deliver high-quality insights and facilitate research and development in healthcare and pharmaceutical innovation. This model is currently being actively updated and improved as part of my ongoing research and work in Retrieval-Augmented Generation (RAG).

🚀 Key Features

Fine-Tuned on: Clinical
Primary Use Case(s): [Summarization, Question Answering, etc.]
Updates in Progress:
- Optimizing for better accuracy with RAG workflows.
- Incorporating new datasets and training strategies.
- Fine-tuning with community feedback.

📅 What's Next?

I am actively working on:

Integrating this model into a RAG pipeline for enhanced retrieval-augmented tasks.
Regular updates to improve performance and reduce inference time.
Expanding support for [languages/domains/etc.].

Stay tuned for updates and improvements in the coming weeks!

🛠️ How to Use

Here's a quick example of how you can use this model:

from transformers import pipeline

# Load the model
model = pipeline("task_name", model="hackint0sh/phi-3-clinical")

# Example usage
input_text = "Your input here"
output = model(input_text)
print(output)

Replace task_name with the appropriate task (e.g., "text-classification", "question-answering", "Clinical Trial Format").

🙋‍♂️ Need Help?

I’m here to help! If you have any questions, suggestions, or encounter any issues while using the model, feel free to: • Open an Issue on this repository. • DM me directly on Hugging Face.

I’m always happy to collaborate and improve this model further based on your feedback. 😊

🌟 Contributing

Contributions are welcome! If you have ideas for improvements or want to contribute, feel free to fork this repository and open a pull request.

📝 License

This model is released under the MIT license. See the LICENSE file for more details.

Thank you for your interest in Phi-3-Clinical! Your support and feedback help make this model better for everyone. ❤️

Downloads last month: 20

Safetensors

Model size

4B params

Tensor type

F16

Model tree for hackint0sh/phi-3-clinical

Finetunes

1 model

Merges

1 model

Quantizations

4 models