Skip to content
LlamaIndex Python Documentation
Search
Ctrl
K
Cancel
TypeScript
Twitter
LinkedIn
Bluesky
GitHub
Select theme
Dark
Light
Auto
LlamaCloud
Welcome to LlamaCloud
Parse
Overview of LlamaParse
Getting Started
Presets and Modes
Advanced Parsing Modes
Auto Mode
Output
Modes and Presets
Features
Parsing options
Multimodal Parsing
Python Usage
Layout Extraction
Metadata
Cache options
Structured Output (Beta)
Webhook
Supported Document Types
Job predictability
Selecting what to parse
LlamaParse Document Pipeline Triggers
Parsing instructions (deprecated)
Prompts
LlamaParse FAQ
LlamaParse API v2 Guide
Examples
LlamaParse Examples
Parse and Analyze Excel Spreadsheets with LlamaParse
Parse with Additional Prompts
Migration Guide: Parse Upload Endpoint v1 to v2
Extract
Getting Started
Getting Started with LlamaExtract
LlamaExtract REST API
LlamaExtract Python SDK
Using the LlamaExtract Web UI
Examples
LlamaExtract Examples
Auto-Generate Schema for Extraction
Extract Data from Financial Reports - with Citations and Reasoning
Features
LlamaExtract Core Concepts
LlamaExtract Extensions
LlamaExtract Configuration Options
LlamaExtract Performance Tips
LlamaExtract Schema Design
LlamaExtract Schema Restrictions
LlamaExtract Privacy
LlamaExtract Pricing and Usage Data
Classify
Examples
LlamaClassify Examples
Classify Contract Types
Getting Started
Getting Started with Classify
Classify Python SDK
Index
Getting Started
Getting Started with LlamaCloud
LlamaCloud Quick Start
Usage Guides
LlamaCloud API & Clients Guide
LlamaCloud Framework Integration
LlamaCloud Usage Guides
LlamaCloud No-code UI Guide
How-to Guides
Files
Extracting Figures from Documents
Getting Started with LlamaCloud Index & Agents
Integrations
Data Sinks
Data Sinks
AstraDB
Azure AI Search
Managed Data Sink
Milvus
MongoDB Atlas Vector Search
Pinecone
Qdrant
Data Sources
LlamaCloud Data Sources
Azure Blob Storage Data Source
Box Storage Data Source
Confluence Data Source
File Upload Data Source
Google Drive Data Source
Jira Data Source
Notion Data Source
Microsoft OneDrive Data Source
S3 Data Source
Microsoft SharePoint Data Source
Slack Data Source
Embedding Models
Embedding Models
Azure Embedding
Bedrock Embedding
Cohere Embedding
Gemini Embedding
HuggingFace Embedding
OpenAI Embedding
Multi-Environments
Parsing & Transformation in LlamaCloud
Retrieval
Advanced
Basic
Composite Retrieval
Image Retrieval
Retrieval Modes
General
Webhooks
API Key
Self-Hosting
Architecture
Basic Setup
Auth
Databases and Queues
Azure Service Bus as Job Queue
Overview
File Storage
Ingress
LLM Integrations
Anthropic API Setup
AWS Bedrock Setup
Azure OpenAI Setup
Google Gemini API Setup
Google Vertex AI Setup
OpenAI Setup
Overview
Frequently Asked Questions
Get Started
Tuning
Service Configurations
Credit Pricing & Usage
Organizations
Cookbooks
Cookbooks
Enterprise Rollout
Regions
Managing Your Subscription
Cloud API Reference 🔗
Workflows
Introduction
Customizing entry and exit points
Deploying a Workflow
Drawing a Workflow
Writing durable workflows
Managing events
Managing State
Resource Objects
Retry steps execution
Workflows API Reference 🔗
LlamaIndex Framework
Welcome to LlamaIndex 🦙 !
Getting Started
High-Level Concepts
Installation and Setup
How to read these docs
Starter Tutorial (Using OpenAI)
Starter Tutorial (Using Local LLMs)
Discover LlamaIndex Video Series
Frequently Asked Questions (FAQ)
Starter Tools
Starter Tools
RAG CLI
Async Programming in Python
Learn
Building an LLM application
Using LLMs
Building agents
Building an agent
Using existing tools
Maintaining state
Streaming output and events
Human in the loop
Multi-agent patterns in LlamaIndex
Using Structured Output
Building Workflows
Workflows introduction
Basic workflow
Branches and loops
Maintaining state
Streaming events
Concurrent execution of workflows
Subclassing workflows
Resources
Observability
Workflows from unbound functions
Building a RAG pipeline
Introduction to RAG
Indexing
Indexing
Loading
Loading Data (Ingestion)
Loading from LlamaCloud
LlamaHub
Querying
Querying
Storing
Storing
Structured Data Extraction
Introduction to Structured Data Extraction
Using Structured LLMs
Structured Prediction
Low-level structured data extraction
Structured Input
Tracing And Debugging
Tracing and Debugging
Evaluating
Cost Analysis
Cost Analysis
Usage Pattern
Evaluating
Putting It All Together
Putting It All Together
Agents
Apps
Full-Stack Web Application
A Guide to Building a Full-Stack Web App with LLamaIndex
A Guide to Building a Full-Stack LlamaIndex Web App with Delphic
Chatbots
How to Build a Chatbot
Q And A
Q&A patterns
A Guide to Extracting Terms and Definitions
Structured Data
Structured Data
Privacy and Security
Use Cases
Use Cases
Agents
Chatbots
Structured Data Extraction
Fine-tuning
Querying Graphs
Multi-modal
Prompting
Question-Answering (RAG)
Querying CSVs
Parsing Tables and Charts
Text to SQL
Component Guides
Component Guides
Deploying
Agents
Agents
Memory
Module Guides
Tools
Chat Engines
Chat Engine
Module Guides
Usage Pattern
Query Engine
Query Engine
Module Guides
Response Modes
Streaming
Supporting Modules
Usage Pattern
Evaluating
Evaluating
Contributing A `LabelledRagDataset`
Evaluating Evaluators with `LabelledEvaluatorDataset`'s
Evaluating With `LabelledRagDataset`'s
Modules
Usage Pattern (Response Evaluation)
Usage Pattern (Retrieval)
Indexing
Indexing
Document Management
How Each Index Works
LlamaCloudIndex + LlamaCloudRetriever
Using a Property Graph Index
Metadata Extraction
Module Guides
Using VectorStoreIndex
Loading
Loading Data
Connector
Data Connectors (LlamaHub)
LlamaParse
Module Guides
Usage Pattern
Documents And Nodes
Documents / Nodes
Defining and Customizing Documents
Metadata Extraction Usage Pattern
Defining and Customizing Nodes
Ingestion Pipeline
Ingestion Pipeline
Transformations
Node Parsers
Node Parser Usage Pattern
Node Parser Modules
SimpleDirectoryReader
MCP
Model Context Protocol (MCP)
Converting Existing LlamaIndex Workflows & Tools to MCP
LlamaCloud MCP Servers & Tools
Using MCP Tools with LlamaIndex
Models
Models
Embeddings
Llms
Using LLMs
Using local models
Available LLM integrations
Customizing LLMs within LlamaIndex Abstractions
Using LLMs as standalone modules
Multi-modal models
Prompts
Prompts
Prompt Usage Pattern
Observability
Observability
Callbacks
Callbacks
Token Counting - Migration Guide
Instrumentation
Querying
Querying
Node Postprocessors
Node Postprocessor
Node Postprocessor Modules
Response Synthesizers
Response Synthesizer
Response Synthesis Modules
Retriever
Retriever
Retriever Modes
Retriever Modules
Router
Routers
Structured Outputs
Structured Outputs
Output Parsing Modules
Pydantic Programs
(Deprecated) Query Engines + Pydantic Outputs
Storing
Storing
Chat Stores
Customizing Storage
Document Stores
Index Stores
Key-Value Stores
Persisting & Loading Data
Vector Stores
Supporting Modules
Migrating from ServiceContext to Settings
Configuring Settings
Supporting Modules
Workflow
Workflows
Open Source Community
FAQ
Frequently Asked Questions
Chat Engines
Documents and Nodes
Embeddings
Large Language Models
Query Engines
Vector Database
Full-Stack Projects
Integrations
Integrations
ChatGPT Plugin Integrations
Unit Testing LLMs/RAG With DeepEval
Fleet Context Embeddings - Building a Hybrid Search Engine for the Llamaindex Library
Using Graph Stores
Tracing with Graphsignal
Guidance
LM Format Enforcer
Using Managed Indices
Tonic Validate
Evaluating and Tracking with TruLens
Perform Evaluations on LlamaIndex with UpTrain
Using Vector Stores
Llama Packs
Llama Packs 🦙📦
ChangeLog
Framework API Reference 🔗
Twitter
LinkedIn
Bluesky
GitHub
Select theme
Dark
Light
Auto
Test delete
if ref_doc_id: store.delete(ref_doc_id) print(store._collection.count_documents({}))
