AnythingLLM REST API — Complete Developer Reference

The AnythingLLM Developer API gives you full programmatic control over your instance — from spinning up workspaces and uploading documents to running multi-turn chats and managing users. Every operation available in the UI is also available through the REST API, making it straightforward to integrate AnythingLLM into your own applications, pipelines, and automation workflows.

Base URL

All API requests are served from your running AnythingLLM server. By default the server listens on port 3001, and every endpoint lives under the /api/v1/ prefix:

http://your-instance:3001/api/v1/

Replace your-instance with the hostname or IP address where AnythingLLM is running. For local development this is typically localhost.

Interactive API Explorer

AnythingLLM ships with a built-in Swagger UI that documents every endpoint and lets you make live requests directly from your browser:

http://your-instance:3001/api/docs

To disable the Swagger UI in production, set the environment variable DISABLE_SWAGGER_DOCS=true in your server’s .env file. This prevents the interactive docs from being publicly accessible without affecting normal API functionality.

Authentication

Every request must include a valid API key in the Authorization header as a Bearer token:

Authorization: Bearer YOUR_API_KEY

Generate an API key from Settings → API Keys → Generate New API Key inside your AnythingLLM instance. See the Authentication page for full details.

General Conventions

Property	Value
Protocol	HTTP / HTTPS
Request format	`application/json`
Response format	`application/json` (all endpoints)
Rate limiting	None by default
Auth scheme	Bearer token (`Authorization` header)
Error shape	`{ "error": "message" }`

All timestamps are returned as ISO 8601 strings. Boolean flags are native JSON booleans. Endpoints that return lists always wrap the array in a named key (e.g. { "workspaces": [...] }).

Quick Start

Use the following curl command to verify your API key is valid and your instance is reachable:

curl -X GET "http://your-instance:3001/api/v1/auth" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json"

A successful response looks like this:

{
  "authenticated": true
}

If the key is missing or invalid, the API returns 403 Forbidden with { "error": "No valid api key found." }.

API Endpoint Groups

Authentication

Verify that an API key is valid. A lightweight health-check for auth before executing more complex requests.

Workspaces

Create, update, and delete workspaces. Run single-turn and streaming chats, perform vector searches, and manage embedded documents.

Documents

Upload files or raw links for processing, list documents in storage, move files between folders, and remove content from your instance.

Admin

Manage users, roles, invites, and system-wide preferences. All admin endpoints require multi-user mode to be enabled.

Embeds

Create and manage embeddable chat widgets. Control chat mode, session limits, domains, and retrieve embed chat history.

OpenAI Compatible

Drop-in replacements for OpenAI’s /models, /chat/completions, and /embeddings endpoints. Point any OpenAI SDK at your AnythingLLM instance.

Endpoint Summary

Authentication

Method	Path	Description
`GET`	`/v1/auth`	Verify a Bearer token is valid

Workspaces

Method	Path	Description
`POST`	`/v1/workspace/new`	Create a new workspace
`GET`	`/v1/workspaces`	List all workspaces
`GET`	`/v1/workspace/{slug}`	Get a workspace by slug
`POST`	`/v1/workspace/{slug}/update`	Update workspace settings
`DELETE`	`/v1/workspace/{slug}`	Delete a workspace
`GET`	`/v1/workspace/{slug}/chats`	Retrieve workspace chat history
`POST`	`/v1/workspace/{slug}/update-embeddings`	Add or remove embedded documents
`POST`	`/v1/workspace/{slug}/update-pin`	Pin or unpin a document
`POST`	`/v1/workspace/{slug}/chat`	Send a chat message
`POST`	`/v1/workspace/{slug}/stream-chat`	Stream a chat response
`POST`	`/v1/workspace/{slug}/vector-search`	Semantic vector search
`POST`	`/v1/workspace/{slug}/thread/new`	Create a workspace thread
`POST`	`/v1/workspace/{slug}/thread/{threadSlug}/update`	Update a thread
`DELETE`	`/v1/workspace/{slug}/thread/{threadSlug}`	Delete a thread
`GET`	`/v1/workspace/{slug}/thread/{threadSlug}/chats`	Get thread chat history
`POST`	`/v1/workspace/{slug}/thread/{threadSlug}/chat`	Send a chat to a thread
`POST`	`/v1/workspace/{slug}/thread/{threadSlug}/stream-chat`	Stream a chat to a thread

Documents

Method	Path	Description
`POST`	`/v1/document/upload`	Upload a file for processing
`POST`	`/v1/document/upload/{folderName}`	Upload a file into a named folder
`POST`	`/v1/document/upload-link`	Upload a URL for scraping
`POST`	`/v1/document/raw-text`	Upload raw text as a document
`GET`	`/v1/documents`	List all documents
`GET`	`/v1/documents/folder/{folderName}`	List documents in a folder
`GET`	`/v1/document/accepted-file-types`	List accepted file types
`GET`	`/v1/document/metadata-schema`	Get metadata key schema
`GET`	`/v1/document/{docName}`	Get a single document
`POST`	`/v1/document/create-folder`	Create a document folder
`DELETE`	`/v1/document/remove-folder`	Remove a document folder
`POST`	`/v1/document/move-files`	Move documents between folders
`GET`	`/v1/document/generated-files/{filename}`	Download an agent-generated file

Admin (multi-user mode only)

Method	Path	Description
`GET`	`/v1/admin/is-multi-user-mode`	Check multi-user mode status
`GET`	`/v1/admin/users`	List all users
`POST`	`/v1/admin/users/new`	Create a new user
`POST`	`/v1/admin/users/{id}`	Update a user
`DELETE`	`/v1/admin/users/{id}`	Delete a user
`GET`	`/v1/admin/invites`	List all invites
`POST`	`/v1/admin/invite/new`	Create an invite link
`DELETE`	`/v1/admin/invite/{id}`	Deactivate an invite
`GET`	`/v1/admin/workspaces/{workspaceId}/users`	List users with workspace access
`POST`	`/v1/admin/workspaces/{workspaceSlug}/manage-users`	Set workspace user permissions
`POST`	`/v1/admin/workspace-chats`	List all system chats (paginated)
`POST`	`/v1/admin/preferences`	Update instance-wide preferences

Embeds

Method	Path	Description
`GET`	`/v1/embed`	List all embed configurations
`POST`	`/v1/embed/new`	Create a new embed
`POST`	`/v1/embed/{embedUuid}`	Update an embed configuration
`DELETE`	`/v1/embed/{embedUuid}`	Delete an embed
`GET`	`/v1/embed/{embedUuid}/chats`	List all chats for an embed
`GET`	`/v1/embed/{embedUuid}/chats/{sessionUuid}`	Get chats for a specific embed session

OpenAI Compatible

Method	Path	Description
`GET`	`/v1/openai/models`	List available workspace models
`POST`	`/v1/openai/chat/completions`	OpenAI-compatible chat completions
`POST`	`/v1/openai/embeddings`	OpenAI-compatible text embeddings
`GET`	`/v1/openai/vector_stores`	List workspaces as vector stores

Overview

Endpoints

AnythingLLM REST API — Complete Developer Reference

Base URL

Interactive API Explorer

Authentication

General Conventions

Quick Start

API Endpoint Groups

Authentication

Workspaces

Documents

Admin

Embeds

OpenAI Compatible

Endpoint Summary

Authentication

Workspaces

Documents

Admin (multi-user mode only)

Embeds

OpenAI Compatible

Build docs developers (and LLMs) love

Overview

Endpoints

Documentation Index

​Base URL

​Interactive API Explorer

​Authentication

​General Conventions

​Quick Start

​API Endpoint Groups

Authentication

Workspaces

Documents

Admin

Embeds

OpenAI Compatible

​Endpoint Summary

​Authentication

​Workspaces

​Documents

​Admin (multi-user mode only)

​Embeds

​OpenAI Compatible

Build docs developers (and LLMs) love

Base URL

Interactive API Explorer

Authentication

General Conventions

Quick Start

API Endpoint Groups

Endpoint Summary

Authentication

Workspaces

Documents

Admin (multi-user mode only)

Embeds

OpenAI Compatible