← All repos

pandas-ai

Chat with your database or your datalake (SQL, CSV, parquet). PandasAI makes data analysis conversational using LLMs and RAG.

GitHubpandas-ai.com
aicsvdatadata-analysisdatabasedatalakedata-sciencedata-visualizationgpt-4llmpandassqltext-to-sql
Browse cluster: SQL databases and distributed systems →
23,746stars
1,416commits
120contributors
3languages

Tech stack & purpose

PandasAI is a Python library that enables natural language interaction with data by allowing users to ask questions about their databases, data lakes, and files in CSV or Parquet format. The project is built around large language models (LLMs) and retrieval-augmented generation (RAG) to make data analysis conversational, supporting multiple data sources including SQL databases like PostgreSQL and MySQL, cloud platforms like Google BigQuery, Snowflake, and Databricks, as well as financial data via Yahoo Finance. The core library is developed by Sinaptik AI and includes optional extensions for various LLM providers like OpenAI and LiteLLM, vector stores such as Pinecone and Qdrant, and a Docker sandbox for secure code execution. PandasAI is available under the MIT license for the core functionality, while enterprise extensions are proprietary.

Community & reference links

Languages

Python
99.4%
Makefile
0.5%
Dockerfile
0.0%

Contributors (top 30 of 120)