LlamaIndex
Document parsing platform plus open-source RAG framework
Overview
LlamaIndex is the San Francisco company behind LlamaParse (formerly LlamaCloud), a credit-priced platform that parses, extracts, classifies, splits and indexes complex documents for AI agents. Its original MIT-licensed Python framework for RAG and agents remains free, but the company says its primary focus has shifted to LlamaParse.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 12 Oct 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
- 5 concurrent parse and extract jobs
- 5 indexes, 50 files per index
- File upload only
- 1 project
- 50 indexes, 500 files per index
- 50 external data sources
- Basic email support
- 20 concurrent jobs
- 100 indexes, 2,000 files per index
- 5 projects
- Priority Slack Connect support
- Enterprise SSO
- SaaS or hybrid deployment and VPC
- Higher rate limits
- Dedicated account manager
Use cases
What you can produce with LlamaIndex
- Parsing of PDFs, Office files, spreadsheets and images (130+ formats) into Markdown or JSON
- Schema-based field extraction with citations and confidence scores
- Classification and splitting of mixed document sets
- Indexing of parsed documents synced from Google Drive, SharePoint, S3 or Confluence
- The free MIT-licensed Python framework for RAG apps and agents
- Local parsing with no cloud calls through open-source LiteParse
- Multi-step document agents with LlamaIndex Workflows
LlamaIndex has moved from being primarily a RAG framework to selling LlamaParse, a document-processing platform for parsing, extraction, classification and indexing, priced in credits from a free plan to custom Enterprise contracts. The MIT-licensed Python framework is still released and free.
It suits teams whose hard problem is turning PDFs, scans, tables and filings into structured data. Developers choosing a general-purpose framework should note the shift in focus, the archived TypeScript port and retired LlamaHub links, and test credit costs on their own documents.
What we found
LlamaIndex began in late 2022 as an open-source Python library for connecting language models to private data, and for most of its life it was reviewed as a RAG framework. That is no longer how the company describes itself. Its homepage now sells AI agents for document OCR and workflows, the commercial platform once called LlamaCloud has been renamed LlamaParse, and in September 2026 the framework's README added that the company's primary focus has shifted towards LlamaParse.
LlamaParse parses, extracts, classifies, splits and indexes documents through one API, handling more than 130 file types including scans, tables, charts and handwriting. Pricing runs on credits at $1.25 per 1,000: a Free plan includes 10,000 a month, Starter costs $50 a month with 40,000, Pro costs $500 a month with 400,000, and Enterprise is custom, adding SSO and SaaS or hybrid deployment. Credit cost per page depends on the parsing tier, so test your own documents before estimating spend. The platform lists SOC 2 Type II, GDPR and HIPAA compliance, with US and EU data regions.
The MIT-licensed framework still ships, with release 0.14.25 in September 2026, and keeps its range of index types, from vector and keyword-table to tree and property-graph, plus hundreds of integration packages. Expect less momentum than before: the TypeScript port is archived on GitHub, and LlamaHub links were removed from the README. Open-source parsing continues separately as LiteParse, an Apache-2.0 local parser.
The fit has changed with the strategy. Teams whose real problem is turning messy PDFs, invoices or filings into reliable structured data are now LlamaIndex's core audience and should trial LlamaParse directly. Teams wanting a general-purpose agent or RAG framework can still use the library at no cost, but should weigh how central it remains to the company's roadmap. LlamaIndex is based in San Francisco and raised a $19 million Series A in March 2025.
Compiled by RECATOOLS from public sources: vendor documentation and pricing, code repositories and published reporting.
About this listing
This entry was compiled from publicly available data including LlamaIndex's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with LlamaIndex unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to LlamaIndex directly →
Spotted something out of date? Suggest an update →
Related reading
Alternatives to LlamaIndex
More in Agents & Automation
