PandasAI

PandasAI is a Python library that executes data analysis code in your environment and lets analysts ask questions in plain language. It can isolate that execution in Docker. Its documented LiteLLM integration uses OpenAI's gpt-4.1-mini, so model requests go to a cloud service and require an API key.

The library works with SQL databases, CSV files and parquet data. Developers can use it to add conversational analysis to data tools, while analysts can ask for calculations without writing the corresponding analysis code themselves. Questions can cover a single dataset or relate multiple pandas DataFrames, such as matching employee records with salary data.

PandasAI can return answers and generate charts from natural-language requests. It combines LLMs with retrieval-augmented generation to support questions about data. The optional Docker sandbox provides an execution boundary for analysis code; the OpenAI integration still relies on remote inference.

The core library uses the MIT expat license, with a separately licensed enterprise directory. PandasAI also offers a managed cloud product and a distinct self-hosted enterprise offering.

The v3.0.0 release is the Python library. Core code uses MIT Expat; the enterprise directory and third-party components retain separate terms. The managed PandasAI website offers a distinct hosted product, so this listing links directly to the library’s source.