metadata

title: SmolLM2 Backend Local Model
emoji: 📊
colorFrom: yellow
colorTo: red
sdk: docker
pinned: false
license: apache-2.0
short_description: Backend of SmolLM2 chatbot with local model
app_port: 7860

SmolLM2 Backend Local Model

This project implements a FastAPI API that uses LangChain and LangGraph to generate text with the Qwen2.5-72B-Instruct model from HuggingFace.

Configuration

In HuggingFace Spaces

This project is designed to run in HuggingFace Spaces. To configure it:

Create a new Space in HuggingFace with SDK Docker
Configure the HUGGINGFACE_TOKEN or HF_TOKEN environment variable in the Space configuration:
- Go to the "Settings" tab of your Space
- Scroll down to the "Repository secrets" section
- Add a new variable with the name HUGGINGFACE_TOKEN and your token as the value
- Save the changes

Local development

For local development:

Clone this repository
Create a .env file in the project root with your HuggingFace token:
```
HUGGINGFACE_TOKEN=your_token_here
```
Install the dependencies:
```
pip install -r requirements.txt
```

Local execution

uvicorn app:app --reload

The API will be available at http://localhost:7860.

Endpoints

GET `/`

Welcome endpoint that returns a greeting message.

POST `/generate`

Endpoint to generate text using the language model.

Request parameters:

{
  "query": "Your question here",
  "thread_id": "optional_thread_identifier"
}

Response:

{
  "generated_text": "Generated text by the model",
  "thread_id": "thread identifier"
}

Docker

To run the application in a Docker container:

# Build the image
docker build -t smollm2-backend .

# Run the container
docker run -p 7860:7860 --env-file .env smollm2-backend

API documentation

The interactive API documentation is available at:

Swagger UI: http://localhost:7860/docs
ReDoc: http://localhost:7860/redoc