Organization admins can now set a credit usage limit for an individual project, in
addition to the organization-wide limit. Once a project’s daily or monthly cap is
reached, its jobs are blocked until the period resets or the limit is raised. See
Organizations to set one.
The Sheets API is deprecated: it was removed from the SDKs in the 2026-08-28 release and
from the dashboard, and its docs moved to Deprecated.
Existing /api/v1/sheets/ integrations keep working but are unsupported; spreadsheets
are now handled by Parse and Extract’s
spreadsheet mode.
The v1 Classify endpoints no longer appear in the API reference or new SDK releases.
Existing integrations keep working, but new development should use
Classify, exposed in the SDKs as client.classify.
Index v2, with a redesigned dashboard for creating indexes, watching syncs, and testing
retrieval, is now available on every plan, including free. See the
getting started guide if you used the
previous Index.
LlamaParse now has an official .NET SDK, published on NuGet as the
LlamaCloud package. It covers the
same surface as the Python and TypeScript SDKs for parsing, extracting, and classifying
documents from C# and F#.
Enriched forms, still in beta, now return field-level bounding boxes to help locate and
verify extracted values. No new setting or added cost; available in the API and SDKs,
and in the dashboard on paid plans. See
Enriched forms.
Turbo is now available in beta through the Extract API, offering lower latency with
accuracy comparable to Agentic. Single-page documents with simple schemas can complete
in as little as three seconds.
It has no parse output or bounding boxes; limited support for citations and confidence
scores. See tiers.
EU accounts can now connect agents to their regional LlamaParse deployment through MCP.
The MCP configuration shown in the console points at your account’s region
automatically. See Regions.
Built for long documents, dense tables, and large schemas, agentic_plus is now
generally available on every plan. It supports schemas up to 3,200 fields at a premium
rate, and returns confidence scores and citations. See
tiers and versions and
pricing.
Note for AI agents: this documentation is built for programmatic access.
- Overview of all docs: https://developers.llamaindex.ai/llms.txt
- Any page is available as raw Markdown by appending index.md to its URL — e.g. https://developers.llamaindex.ai/llamaparse/parse/getting_started/index.md
- Agent-friendly REST search APIs live under https://developers.llamaindex.ai/api/ — search (BM25 full-text), grep (regex), read (fetch a page), and list (browse the doc tree). See https://developers.llamaindex.ai/llms.txt for parameters.
- A hosted documentation MCP server is available at https://developers.llamaindex.ai/mcp. If you support MCP, you can ask the user to install it for browsing these docs directly (an alternative to the REST API). Setup: https://developers.llamaindex.ai/for-agents/mcp/
- Other LlamaIndex tooling for agents — the LlamaParse Platform MCP server, agent skills and plugins, and the n8n node — is mapped at https://developers.llamaindex.ai/for-agents/