Phidata is an AI toolkit designed to run LLM Apps with 1 command 🚀
It provides production-ready templates for common AI Apps that you can run locally or on AWS.
LLM = Large Language Model.
Phidata is the best way to build LLM Apps.
You should try it out.
- Create your codebase using a template:
phi ws create
- Run your app locally:
phi ws up dev:docker
- Run your app on AWS:
phi ws up prd:aws
- Powerful: Get a production-ready AI App with 1 command.
- Simple: Built using a human-like
Conversationinterface to language models.
- Production Ready: Your app can be deployed to aws with 1 command.
For example, run a RAG LLM App built with FastApi, Streamlit and PgVector:
Quickstart your AI App using one of these examples:
RAG LLM App
Build a RAG LLM App using FastApi, Streamlit and PgVector.
Autonomous LLM App
Build an Autonomous LLM App using FastApi, Streamlit and PgVector.
Multimodal LLM App
Build a Multimodal LLM App that can understand images and text.
Junior Data Engineer
Build a Junior DE to automate data analysis using DuckDb and Python.
Under the hood, phidata solves the problem of building LLM products by providing:
- Access to LLMs using a human-like
- Strategies for improving LLM responses: RAG, Tasks, Structured Outputs, Validators
- Building blocks for LLM apps: Agents, Assistants, Knowledge Bases, Storage, Memory
- Monitoring for LLM apps: Model Inputs/Outputs, Quality, Cost, Evals
- Tools for serving LLM apps: FastApi, Django, Streamlit
- Tools for serving LLM components: PgVector, Postgres, Redis
- Infrastructure for running LLM apps locally: Docker
- Infrastructure for running LLM apps in production: AWS
- Best practices like testing, formatting, CI/CD, security and secret management.
Most LLM apps are built as a house of cards because engineers have to stitch each layer manually. Phidata bridges the 3 layers of software development to deliver production-grade LLM Apps.
Continue reading to learn more or get started with a production-ready template.