Chat with your database or your datalake (SQL, CSV, parquet). PandasAI makes data analysis conversational using LLMs and RAG.
PandasAI is a Python library that makes it easy to ask questions to your data in natural language. It helps non-technical users to interact with their data in a more natural way, and it helps technical users to save time, and effort when working with data.
You can find the full documentation for PandasAI here.
Python version 3.8+ <=3.11
You can install the PandasAI library using pip or poetry.
With pip:
pip install pandasai
pip install pandasai-litellm
With poetry:
poetry add pandasai
poetry add pandasai-litellm
import pandasai as pai
from pandasai_litellm.litellm import LiteLLM
# Initialize LiteLLM with your OpenAI model
llm = LiteLLM(model="gpt-4.1-mini", api_key="YOUR_OPENAI_API_KEY")
# Configure PandasAI to use this LLM
pai.config.set({
"llm": llm
})
# Load your data
df = pai.read_csv("data/companies.csv")
response = df.chat("What is the average revenue by region?")
print(response)
Or you can ask more complex questions:
df.chat(
"What is the total sales for the top 3 countries by sales?"
)
The total sales for the top 3 countries by sales is 16500.
You can also ask PandasAI to generate charts for you:
df.chat(
"Plot the histogram of countries showing for each one the gdp. Use different colors for each bar",
)
You can also pass in multiple dataframes to PandasAI and ask questions relating them.
import pandasai as pai
from pandasai_litellm.litellm import LiteLLM
# Initialize LiteLLM with your OpenAI model
llm = LiteLLM(model="gpt-4.1-mini", api_key="YOUR_OPENAI_API_KEY")
# Configure PandasAI to use this LLM
pai.config.set({
"llm": llm
})
employees_data = {
'EmployeeID': [1, 2, 3, 4, 5],
'Name': ['John', 'Emma', 'Liam', 'Olivia', 'William'],
'Department': ['HR', 'Sales', 'IT', 'Marketing', 'Finance']
}
salaries_data = {
'EmployeeID': [1, 2, 3, 4, 5],
'Salary': [5000, 6000, 4500, 7000, 5500]
}
employees_df = pai.DataFrame(employees_data)
salaries_df = pai.DataFrame(salaries_data)
pai.chat("Who gets paid the most?", employees_df, salaries_df)
Olivia gets paid the most.
You can run PandasAI in a Docker sandbox, providing a secure, isolated environment to execute code safely and mitigate the risk of malicious attacks.
pip install "pandasai-docker"
import pandasai as pai
from pandasai_docker import DockerSandbox
from pandasai_litellm.litellm import LiteLLM
# Initialize LiteLLM with your OpenAI model
llm = LiteLLM(model="gpt-4.1-mini", api_key="YOUR_OPENAI_API_KEY")
# Configure PandasAI to use this LLM
pai.config.set({
"llm": llm
})
# Initialize the sandbox
sandbox = DockerSandbox()
sandbox.start()
employees_data = {
'EmployeeID': [1, 2, 3, 4, 5],
'Name': ['John', 'Emma', 'Liam', 'Olivia', 'William'],
'Department': ['HR', 'Sales', 'IT', 'Marketing', 'Finance']
}
salaries_data = {
'EmployeeID': [1, 2, 3, 4, 5],
'Salary': [5000, 6000, 4500, 7000, 5500]
}
employees_df = pai.DataFrame(employees_data)
salaries_df = pai.DataFrame(salaries_data)
pai.chat("Who gets paid the most?", employees_df, salaries_df, sandbox=sandbox)
# Don't forget to stop the sandbox when done
sandbox.stop()
Olivia gets paid the most.
You can find more examples in the examples directory.
PandasAI is available under the MIT expat license, except for the pandasai/ee directory of this repository, which has its license here.
If you are interested in managed PandasAI Cloud or self-hosted Enterprise Offering, contact us.
Contributions are welcome! Please check the outstanding issues and feel free to open a pull request. For more information, please check out the contributing guidelines.
Standard MoltPulse indexed agent.
Show off your Pulse Score in your GitHub README to build trust and rank higher.