Hi HN, we're Ashpreet, Eli and Yash and we're excited to share Phidata: a collection of AI Apps built with open-source tools.<p>While helping teams build AI products, we built templates for spinning up LLM Apps quickly. Today we're open-sourcing our templates for building:
- RAG LLM Apps
- Autonomous LLM Apps
- Multimodal LLM Apps
- Data Engineering LLM Apps<p>Templates are built with FastApi for serving, Streamlit for prototyping, PgVector for vectors and PosgreSQL for storage. Run them locally using docker and in production on AWS - with 1 command.<p>- Github: <a href="https://github.com/phidatahq/phidata">https://github.com/phidatahq/phidata</a><p>- Documentation: <a href="https://docs.phidata.com" rel="nofollow noreferrer">https://docs.phidata.com</a><p>If you're building an AI product, please give it a try and share your feedback about the stack, documentation and developer experience. Your feedback will help us make them better.<p>PS: For a limited time, GPT-4 access is also included with these templates, so you can run them without an API KEY. This is because we provide access through a proxy for our customers but haven't had time to build in authentication yet. While we work on auth -- GPT4 access is included so give it a spin :)
I.have used phidata toolkit for data and AI workstreams at a job and my own startup as a paid customer. I.must say the tools are great, helps productivity and saving infra cost usecases. They have very little operational headache and just works,<p>We are exploring more ways to.use these tools