LlamaParse
LlamaParse is a cloud-based document parsing service that converts complex documents into structured, LLM-ready data. It solves the problem of extracting clean text and tables from PDFs, slides, and other file types for use in RAG and AI applications. It is for developers building AI applications that need to ingest and process documents.
✨ Key features
- Parses PDFs, slides, and other complex documents
- Extracts tables and text into structured data
- Provides LLM-ready output for RAG pipelines
- Offers cloud-based API for scalable processing
- Supports Python and TypeScript SDKs
🎯 Use cases
- Building RAG pipelines that ingest PDF reports
- Extracting data from slide decks for analysis
- Converting scanned documents into searchable text
- Preprocessing documents for LLM fine-tuning
📦 Installation
🧰 Requirements: Requires an API key from LlamaCloud; Python 3.x or Node.js for SDKs.
pip install llama-cloud>=1.0
For TypeScript:
npm install @llamaindex/llama-cloud
🚀 Usage
from llama_cloud import LlamaParse
parser = LlamaParse(api_key='your-api-key')
docs = parser.load_data('file.pdf')
⚠️ Good to know
This repository is deprecated and will be maintained until May 1, 2026; users should migrate to the new packages.
❓ FAQ
Is LlamaParse still supported?
The current repository is deprecated and will be maintained until May 1, 2026. You should migrate to the new packages for continued support.
What are the new packages?
For Python, use pip install llama-cloud>=1.0; for TypeScript, use npm install @llamaindex/llama-cloud.
What file types can LlamaParse handle?
It can parse PDFs, slides, and other complex documents, extracting tables and text.
Do I need an API key?
Yes, you need an API key from LlamaCloud to use the service.
📊 Repository
🤖 Overview, features, install steps and FAQ were generated from the project's README on Sep 4, 2026. Always check the original source before running commands.