Deployment

Our focus is LLM fundamentals, not deployment. Here's a brief overview of how to share your model.

Gradio Web App

creates instant web UIs for ML models:

1import gradio as gr
2
3def calculate(question: str) -> str:
4    return solve(model, tokenizer, question)
5
6demo = gr.Interface(
7    fn=calculate,
8    inputs=gr.Textbox(label="Math Problem", placeholder="two plus three"),
9    outputs=gr.Textbox(label="Result"),
10    title="Calculator Transformer",
11)
12demo.launch()

GitHub Actions → Hugging Face

The repo includes a GitHub Actions workflow that automatically deploys to on every push:

# .github/workflows/deploy.yml
name: Deploy to HF Spaces
on:
  push:
    branches: [main]

jobs:
  deploy:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - run: |
          git push https://huggingface.co/spaces/${{ secrets.HF_USERNAME }}/calculator main