Building a Local GenAI Application using TinyLlama
The other day I thought of building a very simple and minimal GenAI chat application, using Python only. Well naturally, FastAPI was my premier choice for creating the backend and serving the model. As for keeping it minimal, what better to choose than Streamlit.
Now that the choice for serving the model and client was decided, the only thing that remained was,
“Which model do I use? Do I go for some API?”
Read more...