DataFrameIt¶

Enrich DataFrames with LLMs in a simple and structured way

What is it?¶

DataFrameIt processes text in DataFrames using Large Language Models (LLMs) and extracts structured information defined by Pydantic models. One function, one model, one prompt — done.

from pydantic import BaseModel
from typing import Literal
import pandas as pd
from dataframeit import dataframeit

class Sentiment(BaseModel):
    sentiment: Literal['positive', 'negative', 'neutral']
    confidence: Literal['high', 'medium', 'low']

df = pd.DataFrame({'text': ['Excellent product!', 'Terrible service.']})
result = dataframeit(df, Sentiment, "Analyze the sentiment of the text.", text_column='text')

Features¶

Multiple Providers¶

Google Gemini, OpenAI GPT-5, Anthropic Claude 4.5, Cohere, Mistral — all via LangChain.

Structured Output¶

Automatic validation with Pydantic. Define fields, types, and descriptions — the LLM respects them.

Resilience¶

Automatic retry with exponential backoff. Configurable rate limiting. Never lose progress.

Performance¶

Parallel processing with auto-adjustment. Real-time throughput metrics.

Web Search¶

Tavily integration to enrich data with information from the internet.

Multiple Inputs¶

DataFrame, Series, list, dictionary — everything works. Polars included.

Quick Installation¶

pip install dataframeit[google]  # Google Gemini 3 (recommended)
pip install dataframeit[openai]  # OpenAI GPT-5
pip install dataframeit[anthropic]  # Anthropic Claude 4.5

Next Steps¶

Installation ¶

Set up with your preferred provider

Quickstart ¶

First project in 5 minutes

Guides ¶

Parallelism, retry, web search

LLM Reference ¶

Compact docs for code assistants