# Table of Contents - [Home ~ AnythingLLM](#home-anythingllm) - [Roadmap ~ AnythingLLM](#roadmap-anythingllm) - [All Features ~ AnythingLLM](#all-features-anythingllm) - [AI Agents ~ AnythingLLM](#ai-agents-anythingllm) - [Clarifying Questions ~ AnythingLLM](#clarifying-questions-anythingllm) - [Authenticated Scraping ~ AnythingLLM](#authenticated-scraping-anythingllm) - [Chat Modes ~ AnythingLLM](#chat-modes-anythingllm) - [API Access & Keys ~ AnythingLLM](#api-access-keys-anythingllm) - [Appearance Customization ~ AnythingLLM](#appearance-customization-anythingllm) - [What is AnythingLLM ~ AnythingLLM](#what-is-anythingllm-anythingllm) - [Workspace Chat Logs ~ AnythingLLM](#workspace-chat-logs-anythingllm) - [Embedded Chat Widgets ~ AnythingLLM](#embedded-chat-widgets-anythingllm) - [Event Logs ~ AnythingLLM](#event-logs-anythingllm) - [Embedding Models ~ AnythingLLM](#embedding-models-anythingllm) - [Large Language Models ~ AnythingLLM](#large-language-models-anythingllm) - [Transcription Models ~ AnythingLLM](#transcription-models-anythingllm) - [Vector Databases ~ AnythingLLM](#vector-databases-anythingllm) - [Security and Access ~ AnythingLLM](#security-and-access-anythingllm) - [Privacy & Data ~ AnythingLLM](#privacy-data-anythingllm) - [AnythingLLM Default Embedder ~ AnythingLLM](#anythingllm-default-embedder-anythingllm) - [What is the Model Router? ~ AnythingLLM](#what-is-the-model-router-anythingllm) - [LM Studio Embedder ~ AnythingLLM](#lm-studio-embedder-anythingllm) - [Azure OpenAI Embedder ~ AnythingLLM](#azure-openai-embedder-anythingllm) - [System Prompt Variables ~ AnythingLLM](#system-prompt-variables-anythingllm) - [Local AI Embedder ~ AnythingLLM](#local-ai-embedder-anythingllm) - [Overview ~ AnythingLLM](#overview-anythingllm) - [Memories & Personalization ~ AnythingLLM](#memories-personalization-anythingllm) - [Cohere Embedder ~ AnythingLLM](#cohere-embedder-anythingllm) - [OpenAI Embedder ~ AnythingLLM](#openai-embedder-anythingllm) - [AnythingLLM Default LLM ~ AnythingLLM](#anythingllm-default-llm-anythingllm) - [LMStudio LLM ~ AnythingLLM](#lmstudio-llm-anythingllm) - [Ollama Embedder ~ AnythingLLM](#ollama-embedder-anythingllm) - [Local AI LLM ~ AnythingLLM](#local-ai-llm-anythingllm) - [KobaldCPP LLM ~ AnythingLLM](#kobaldcpp-llm-anythingllm) - [Ollama LLM ~ AnythingLLM](#ollama-llm-anythingllm) - [Anthropic LLM ~ AnythingLLM](#anthropic-llm-anythingllm) - [Google Gemini LLM ~ AnythingLLM](#google-gemini-llm-anythingllm) - [Azure OpenAI LLM ~ AnythingLLM](#azure-openai-llm-anythingllm) - [APIpie ~ AnythingLLM](#apipie-anythingllm) - [Transcription Models ~ AnythingLLM](#transcription-models-anythingllm) - [OpenAI Transcription Model ~ AnythingLLM](#openai-transcription-model-anythingllm) - [General Desktop Information ~ AnythingLLM](#general-desktop-information-anythingllm) - [Debug ~ AnythingLLM](#debug-anythingllm) - [LLM Instruction ~ AnythingLLM](#llm-instruction-anythingllm) - [What is an Agent Flow? ~ AnythingLLM](#what-is-an-agent-flow-anythingllm) - [Read File ~ AnythingLLM](#read-file-anythingllm) - [General Help ~ AnythingLLM](#general-help-anythingllm) - [Automatic document sync ~ AnythingLLM](#automatic-document-sync-anythingllm) - [v1.15.0 ~ AnythingLLM](#v1-15-0-anythingllm) - [SQL Agent ~ AnythingLLM](#sql-agent-anythingllm) - [Desktop Changelog Overview ~ AnythingLLM](#desktop-changelog-overview-anythingllm) - [v1.14.2 ~ AnythingLLM](#v1-14-2-anythingllm) - [Contribute ~ AnythingLLM](#contribute-anythingllm) - [v1.7.0 ~ AnythingLLM](#v1-7-0-anythingllm) - [v1.8.4 ~ AnythingLLM](#v1-8-4-anythingllm) - [v1.7.7 ~ AnythingLLM](#v1-7-7-anythingllm) - [v1.7.1 ~ AnythingLLM](#v1-7-1-anythingllm) - [v1.9.1 ~ AnythingLLM](#v1-9-1-anythingllm) - [Support ~ AnythingLLM](#support-anythingllm) - [v1.8.2 ~ AnythingLLM](#v1-8-2-anythingllm) - [v1.8.1 ~ AnythingLLM](#v1-8-1-anythingllm) - [v1.8.0 ~ AnythingLLM](#v1-8-0-anythingllm) - [v1.7.5 ~ AnythingLLM](#v1-7-5-anythingllm) - [v1.7.6 ~ AnythingLLM](#v1-7-6-anythingllm) - [v1.6.7 ~ AnythingLLM](#v1-6-7-anythingllm) - [v1.6.10 ~ AnythingLLM](#v1-6-10-anythingllm) - [v1.6.6 ~ AnythingLLM](#v1-6-6-anythingllm) - [OpenAI (Generic) LLM ~ AnythingLLM](#openai-generic-llm-anythingllm) - [v1.6.3 ~ AnythingLLM](#v1-6-3-anythingllm) - [Create Scheduled Jobs ~ AnythingLLM](#create-scheduled-jobs-anythingllm) - [TrueFoundry AI gateway ~ AnythingLLM](#truefoundry-ai-gateway-anythingllm) - [Chroma Vector Database ~ AnythingLLM](#chroma-vector-database-anythingllm) - [Vector Databases ~ AnythingLLM](#vector-databases-anythingllm) - [ChatUI Walkthrough ~ AnythingLLM](#chatui-walkthrough-anythingllm) - [Getting Your Pro Key ~ AnythingLLM](#getting-your-pro-key-anythingllm) - [Importing from the AnythingLLM Community Hub ~ AnythingLLM](#importing-from-the-anythingllm-community-hub-anythingllm) - [System Requirements ~ AnythingLLM](#system-requirements-anythingllm) - [Linux Installation ~ AnythingLLM](#linux-installation-anythingllm) - [FAQ ~ AnythingLLM](#faq-anythingllm) - [Uploading to the AnythingLLM Community Hub ~ AnythingLLM](#uploading-to-the-anythingllm-community-hub-anythingllm) - [Update ~ AnythingLLM](#update-anythingllm) - [Uninstall ~ AnythingLLM](#uninstall-anythingllm) - [Installation Overview ~ AnythingLLM](#installation-overview-anythingllm) - [Available Images ~ AnythingLLM](#available-images-anythingllm) - [System Requirements ~ AnythingLLM](#system-requirements-anythingllm) - [Quickstart ~ AnythingLLM](#quickstart-anythingllm) - [502 Error on AnythingLLM Hosted ~ AnythingLLM](#502-error-on-anythingllm-hosted-anythingllm) - [Introduction ~ AnythingLLM](#introduction-anythingllm) - [AI Agents ~ AnythingLLM](#ai-agents-anythingllm) - [API Call ~ AnythingLLM](#api-call-anythingllm) - [AI Agent Setup ~ AnythingLLM](#ai-agent-setup-anythingllm) - [Intelligent Tool Selection ~ AnythingLLM](#intelligent-tool-selection-anythingllm) - [File System Agent ~ AnythingLLM](#file-system-agent-anythingllm) - [Introduction to custom agent skills ~ AnythingLLM](#introduction-to-custom-agent-skills-anythingllm) - [AnythingLLM Browser Extension ~ AnythingLLM](#anythingllm-browser-extension-anythingllm) - [Introduction ~ AnythingLLM](#introduction-anythingllm) - [Features ~ AnythingLLM](#features-anythingllm) - [v1.14.0 ~ AnythingLLM](#v1-14-0-anythingllm) - [Manual QNN Model Download ~ AnythingLLM](#manual-qnn-model-download-anythingllm) - [v1.14.1 ~ AnythingLLM](#v1-14-1-anythingllm) - [v1.12.1 ~ AnythingLLM](#v1-12-1-anythingllm) - [v1.12.0 ~ AnythingLLM](#v1-12-0-anythingllm) - [v1.11.2 ~ AnythingLLM](#v1-11-2-anythingllm) - [v1.11.0 ~ AnythingLLM](#v1-11-0-anythingllm) - [v1.8.5 ~ AnythingLLM](#v1-8-5-anythingllm) - [Why AnythingLLM Pro? ~ AnythingLLM](#why-anythingllm-pro-anythingllm) - [Managing Your Subscription ~ AnythingLLM](#managing-your-subscription-anythingllm) - [MacOS Installation ~ AnythingLLM](#macos-installation-anythingllm) - [Desktop Installation Overview ~ AnythingLLM](#desktop-installation-overview-anythingllm) - [Local Docker Installation ~ AnythingLLM](#local-docker-installation-anythingllm) - [AnythingLLM Cloud Terms & Conditions ~ AnythingLLM](#anythingllm-cloud-terms-conditions-anythingllm) - [Scheduled Jobs ~ AnythingLLM](#scheduled-jobs-anythingllm) - [AnythingLLM's Desktop Assistant ~ AnythingLLM](#anythingllm-s-desktop-assistant-anythingllm) - [AnythingLLM Cloud Privacy Policy ~ AnythingLLM](#anythingllm-cloud-privacy-policy-anythingllm) - [Using Documents in AnythingLLM ~ AnythingLLM](#using-documents-in-anythingllm-anythingllm) - [Why is my AI Agent not using tools! ~ AnythingLLM](#why-is-my-ai-agent-not-using-tools-anythingllm) - [Why does the LLM not use my documents ~ AnythingLLM](#why-does-the-llm-not-use-my-documents-anythingllm) - [v1.11.1 ~ AnythingLLM](#v1-11-1-anythingllm) - ['Fetch failed' error on embed ~ AnythingLLM](#-fetch-failed-error-on-embed-anythingllm) - [v1.6.8 ~ AnythingLLM](#v1-6-8-anythingllm) - [v1.7.3 ~ AnythingLLM](#v1-7-3-anythingllm) - [Groq LLM ~ AnythingLLM](#groq-llm-anythingllm) - [v1.8.3 ~ AnythingLLM](#v1-8-3-anythingllm) - [v1.7.4 ~ AnythingLLM](#v1-7-4-anythingllm) - [Document Generation Agent ~ AnythingLLM](#document-generation-agent-anythingllm) - [Magic Echo ~ AnythingLLM](#magic-echo-anythingllm) - [Privacy Policy ~ AnythingLLM](#privacy-policy-anythingllm) - [MCP on AnythingLLM Desktop ~ AnythingLLM](#mcp-on-anythingllm-desktop-anythingllm) - [handler.js reference ~ AnythingLLM](#handler-js-reference-anythingllm) - [AI Agent Usage ~ AnythingLLM](#ai-agent-usage-anythingllm) - [Telegram Bot ~ AnythingLLM](#telegram-bot-anythingllm) - [Tutorial: HackerNews Flow ~ AnythingLLM](#tutorial-hackernews-flow-anythingllm) - [Custom Agent Skill Developer Guide ~ AnythingLLM](#custom-agent-skill-developer-guide-anythingllm) - [v1.7.2 ~ AnythingLLM](#v1-7-2-anythingllm) - [v1.6.9 ~ AnythingLLM](#v1-6-9-anythingllm) - [v1.10.0 ~ AnythingLLM](#v1-10-0-anythingllm) - [Magic Tab ~ AnythingLLM](#magic-tab-anythingllm) - [Import an LLM into AnythingLLM ~ AnythingLLM](#import-an-llm-into-anythingllm-anythingllm) - [Magic Beacon ~ AnythingLLM](#magic-beacon-anythingllm) - [Cloud Docker Installation ~ AnythingLLM](#cloud-docker-installation-anythingllm) - [Windows Installation ~ AnythingLLM](#windows-installation-anythingllm) - [MCP Compatibility in AnythingLLM ~ AnythingLLM](#mcp-compatibility-in-anythingllm-anythingllm) - [Setting up a Model Router ~ AnythingLLM](#setting-up-a-model-router-anythingllm) - [MCP on AnythingLLM Docker ~ AnythingLLM](#mcp-on-anythingllm-docker-anythingllm) - [Scheduling & The Cron Builder ~ AnythingLLM](#scheduling-the-cron-builder-anythingllm) - [Viewing Runs & Results ~ AnythingLLM](#viewing-runs-results-anythingllm) - [Creating Your First Job ~ AnythingLLM](#creating-your-first-job-anythingllm) - [Configuration & Limits ~ AnythingLLM](#configuration-limits-anythingllm) - [AI Computer use ~ AnythingLLM](#ai-computer-use-anythingllm) - [Features ~ AnythingLLM](#features-anythingllm) - [v1.13.0 ~ AnythingLLM](#v1-13-0-anythingllm) - [Google Calendar Agent ~ AnythingLLM](#google-calendar-agent-anythingllm) - [Privacy Policy ~ AnythingLLM](#privacy-policy-anythingllm) - [Terms of Use ~ AnythingLLM](#terms-of-use-anythingllm) - [Terms of Use ~ AnythingLLM](#terms-of-use-anythingllm) - [plugin.json reference ~ AnythingLLM](#plugin-json-reference-anythingllm) - [v1.9.0 ~ AnythingLLM](#v1-9-0-anythingllm) - [oMLX LLM ~ AnythingLLM](#omlx-llm-anythingllm) - [Cohere LLM ~ AnythingLLM](#cohere-llm-anythingllm) - [Gmail Agent ~ AnythingLLM](#gmail-agent-anythingllm) - [Outlook Agent ~ AnythingLLM](#outlook-agent-anythingllm) - [HuggingFace LLM ~ AnythingLLM](#huggingface-llm-anythingllm) - [OpenAI LLM ~ AnythingLLM](#openai-llm-anythingllm) - [OpenRouter LLM ~ AnythingLLM](#openrouter-llm-anythingllm) - [Perplexity AI LLM ~ AnythingLLM](#perplexity-ai-llm-anythingllm) - [Together AI LLM ~ AnythingLLM](#together-ai-llm-anythingllm) - [AnythingLLM Default Transcription Model ~ AnythingLLM](#anythingllm-default-transcription-model-anythingllm) - [Lance DB Vector Database ~ AnythingLLM](#lance-db-vector-database-anythingllm) - [Milvus Vector Database ~ AnythingLLM](#milvus-vector-database-anythingllm) - [AstraDB Vector Database ~ AnythingLLM](#astradb-vector-database-anythingllm) - [Pinecone Vector Database ~ AnythingLLM](#pinecone-vector-database-anythingllm) - [QDrant Vector Database ~ AnythingLLM](#qdrant-vector-database-anythingllm) - [Weaviate Vector Database ~ AnythingLLM](#weaviate-vector-database-anythingllm) - [Zilliz Vector Database ~ AnythingLLM](#zilliz-vector-database-anythingllm) - [What is the Community Hub? ~ AnythingLLM](#what-is-the-community-hub-anythingllm) - [Debug ~ AnythingLLM](#debug-anythingllm) - [AnythingLLM Cloud ~ AnythingLLM](#anythingllm-cloud-anythingllm) - [RAG Search ~ AnythingLLM](#rag-search-anythingllm) - [Web Browsing ~ AnythingLLM](#web-browsing-anythingllm) - [Save Files ~ AnythingLLM](#save-files-anythingllm) - [Web Scraping ~ AnythingLLM](#web-scraping-anythingllm) - [List Documents ~ AnythingLLM](#list-documents-anythingllm) - [Summarize Documents ~ AnythingLLM](#summarize-documents-anythingllm) - [Chart Generation ~ AnythingLLM](#chart-generation-anythingllm) - [Enable feature previews ~ AnythingLLM](#enable-feature-previews-anythingllm) - [v1.6.4 ~ AnythingLLM](#v1-6-4-anythingllm) - [v1.6.0 ~ AnythingLLM](#v1-6-0-anythingllm) - [v1.6.11 ~ AnythingLLM](#v1-6-11-anythingllm) - [v1.6.1 ~ AnythingLLM](#v1-6-1-anythingllm) - [AWS Bedrock LLM ~ AnythingLLM](#aws-bedrock-llm-anythingllm) - [v1.6.2 ~ AnythingLLM](#v1-6-2-anythingllm) - [Web Scraper ~ AnythingLLM](#web-scraper-anythingllm) - [Write File ~ AnythingLLM](#write-file-anythingllm) - [Debugging flows ~ AnythingLLM](#debugging-flows-anythingllm) - [Configuration ~ AnythingLLM](#configuration-anythingllm) - [AnythingLLM Beta Previews ~ AnythingLLM](#anythingllm-beta-previews-anythingllm) - [Limitations ~ AnythingLLM](#limitations-anythingllm) - [A note about localhost ~ AnythingLLM](#a-note-about-localhost-anythingllm) - [v1.6.5 ~ AnythingLLM](#v1-6-5-anythingllm) - [Overview ~ AnythingLLM](#overview-anythingllm) - [v1.7.8 ~ AnythingLLM](#v1-7-8-anythingllm) - [Getting Started with Flows ~ AnythingLLM](#getting-started-with-flows-anythingllm) - [Mistral AI LLM ~ AnythingLLM](#mistral-ai-llm-anythingllm) - [Overview ~ AnythingLLM](#overview-anythingllm) - [Default Blocks ~ AnythingLLM](#default-blocks-anythingllm) --- # Home ~ AnythingLLM [🚀 AnythingLLM v1.15.0 is live! Update now →](https://anythingllm.com/download) AnythingLLM Documentation ========================= Learn about AnythingLLM's features and how to use them [![AnythingLLM Get Started](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Fgetting-started%2Fheader-image.png&w=3840&q=100)Get Started→](https://docs.anythingllm.com/introduction) [![AnythingLLM Installation](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Fgetting-started%2Finstallation%2Fheader-image.png&w=3840&q=100)Installation→](https://docs.anythingllm.com/installation-desktop/overview) [![AnythingLLM Features](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fheader-image.png&w=3840&q=100)Features→](https://docs.anythingllm.com/features/all-features) [![AnythingLLM Cloud](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Fcloud%2Fheader-image.png&w=3840&q=100)AnythingLLM Cloud→](https://docs.anythingllm.com/cloud/overview) [![AnythingLLM Roadmap](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Fproduct%2Froadmap%2Fheader-image.png&w=3840&q=100)Roadmap→](https://docs.anythingllm.com/roadmap) [![AnythingLLM Changelog](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Fproduct%2Fchangelog%2Fheader-image.png&w=3840&q=100)Changelog→](https://docs.anythingllm.com/changelog/overview) Last updated on June 24, 2026 [AnythingLLM Roadmap](https://docs.anythingllm.com/roadmap "AnythingLLM Roadmap") --- # Roadmap ~ AnythingLLM [🚀 AnythingLLM v1.15.0 is live! Update now →](https://anythingllm.com/download) AnythingLLM Roadmap The Roadmap =========== At AnythingLLM, we're dedicated to making the most advanced LLM application available to everyone. We want to empower everyone to be able leverage LLMs for their own use for both non-technical and technical users. With the community's help, we're making progress towards our goals and this roadmap is a guide for what we're working on and what we plan to work on. This roadmap is not set in stone and is subject to change. We welcome any feedback and suggestions from the community as that is what ultimately drives the product roadmap. **This is for tracking high-level features. Other tracking is done via our [GitHub issues (opens in a new tab)](https://github.com/mintplex-labs/anything-llm/issues) .** * * * * [x] \= Completed * \[~\] = In Progress * [ ] \= Planned _we are working on a public 2026 roadmap and will update this page soon_ Last updated on June 24, 2026 [▲ Home](https://docs.anythingllm.com/ "▲ Home") [Introduction](https://docs.anythingllm.com/introduction "Introduction") --- # All Features ~ AnythingLLM [🚀 AnythingLLM v1.15.0 is live! Update now →](https://anythingllm.com/download) ![AnythingLLM](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fheader-image.png&w=3840&q=100) All Features of AnythingLLM =========================== Click the below cards to know more about the features [![AnythingLLM AI Agents](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fai-agents%2Fheader-image.png&w=3840&q=100)AI Agents→](https://docs.anythingllm.com/features/ai-agents) [![AnythingLLM API Access & Keys](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fapi%2Fheader-image.png&w=3840&q=100)API Access & Keys→](https://docs.anythingllm.com/features/api) [![AnythingLLM Appearance Customization](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fcustomization%2Fheader-image.png&w=3840&q=100)Appearance Customization→](https://docs.anythingllm.com/features/customization) [![AnythingLLM Chat Logs](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fchat-logs%2Fheader-image.png&w=3840&q=100)Chat Logs→](https://docs.anythingllm.com/features/chat-logs) [![AnythingLLM Chat Modes](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fchat-logs%2Fheader-image.png&w=3840&q=100)Chat Modes→](https://docs.anythingllm.com/features/chat-modes) [![AnythingLLM Embedded Chat Widgets](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fchat-widgets%2Fheader-image.png&w=3840&q=100)Embedded Chat Widgets→](https://docs.anythingllm.com/features/chat-widgets) [![AnythingLLM Event Logs](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fevent-logs%2Fheader-image.png&w=3840&q=100)Event Logs→](https://docs.anythingllm.com/features/event-logs) [![AnythingLLM Large Language Models](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Flanguage-models%2Fheader-image.png&w=3840&q=100)Large Language Models→](https://docs.anythingllm.com/features/language-models) [![AnythingLLM Embedding Models](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fembedding-models%2Fheader-image.png&w=3840&q=100)Embedding Models→](https://docs.anythingllm.com/features/embedding-models) [![AnythingLLM Transcription Models](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Ftranscription-models%2Fheader-image.png&w=3840&q=100)Transcription Models→](https://docs.anythingllm.com/features/transcription-models) [![AnythingLLM Vector Database](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fvector-databases%2Fheader-image.png&w=3840&q=100)Vector Databases→](https://docs.anythingllm.com/features/vector-databases) [![AnythingLLM Security & Access](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fsecurity-and-access%2Fheader-image.png&w=3840&q=100)Security & Access→](https://docs.anythingllm.com/features/security-and-access) [![AnythingLLM Privacy & Data Handling](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fprivacy-and-data-handling%2Fheader-image.png&w=3840&q=100)Privacy & Data Handling→](https://docs.anythingllm.com/features/privacy-and-data-handling) [![AnythingLLM Cloud Deployment](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Fgetting-started%2Finstallation%2Fcloud-docker%2Fheader-image.png&w=3840&q=100)Cloud Deployment→](https://docs.anythingllm.com/installation-docker/cloud-docker) [![AnythingLLM System Prompt Variables](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Ftranscription-models%2Fheader-image.png&w=3840&q=100)System Prompt Variables→](https://docs.anythingllm.com/features/system-prompt-variables) [![AnythingLLM Memories](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fmemories%2Fheader-image.png&w=3840&q=100)Memories→](https://docs.anythingllm.com/features/memories) [![AnythingLLM Model Router](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fapi%2Fheader-image.png&w=3840&q=100)Model Router→](https://docs.anythingllm.com/model-router/overview) [Introduction](https://docs.anythingllm.com/introduction "Introduction") [AI Agents](https://docs.anythingllm.com/features/ai-agents "AI Agents") --- # AI Agents ~ AnythingLLM [🚀 AnythingLLM v1.15.0 is live! Update now →](https://anythingllm.com/download) Feature Overview AI Agents ![AnythingLLM AI Agents](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fai-agents%2Fheader-image.png&w=3840&q=100) AI Agents ========= Agents are basically an LLM that has access to some simple tools. We will be adding much more customization in this area soon. All agents share the same tools across workspaces, but operate within the workspace they were invoked via `@agent`. You can start an agent session by going into any workspace and typing `@agent ` and exit by just typing `exit` Agents can scrape websites, list and summarize your documents, search the web, make charts, and even save files to desktop and their own memory. 💡 **Examples:** 1: `@agent` what documents can you see - > LLM will "look" at what are the documents it can see 2: `@agent` summarize readme.pdf - > LLM will summarize that specific embedded file ![AnythingLLM AI Agents](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fai-agents%2Fai-agent.png&w=3840&q=100) [**View all the available `@agent` skills →**](https://docs.anythingllm.com/agent/usage/overview) Last updated on June 24, 2026 [All Features](https://docs.anythingllm.com/features/all-features "All Features") [Agent Surveys](https://docs.anythingllm.com/features/agent-surveys "Agent Surveys") --- # Clarifying Questions ~ AnythingLLM [🚀 AnythingLLM v1.15.0 is live! Update now →](https://anythingllm.com/download) Feature Overview Agent Surveys Clarifying Questions ==================== When an agent receives an ambiguous or incomplete prompt, it typically has to guess at the missing details. The **Clarifying Questions** skill gives your agent a structured way to pause mid-turn and ask the user for the information it needs before continuing — no need to restart the conversation. This opt-in agent skill presents an interactive survey card directly in the chat. The agent decides when to invoke it and what to ask, then resumes with the user's answers fully in context. Why use Clarifying Questions?[](https://docs.anythingllm.com/features/agent-surveys#why-use-clarifying-questions) ------------------------------------------------------------------------------------------------------------------ * **More accurate results** — Instead of guessing, the agent gets the exact details it needs before acting. * **Fewer back-and-forth messages** — The agent can batch multiple independent questions into a single survey card. * **Structured input** — Supports free-form text, URLs, numbers, dates, emails, and single/multi-select choice lists so answers are clean and unambiguous. * **Persistent context** — Answers are saved with the chat history, so both the agent and normal chat can reference them in follow-up turns. 💡 This skill is **off by default** and must be enabled by an admin. It only works in the desktop/browser UI (websocket sessions) — it is not available for API or programmatic agent runs. Enabling the Skill[](https://docs.anythingllm.com/features/agent-surveys#enabling-the-skill) --------------------------------------------------------------------------------------------- 1. Navigate to **Settings → Agent Skills** in the admin sidebar. 2. Under agent skills, click on the slider icon in the top right corner to open agent skill settings modal. 3. Toggle on **Allow agent to ask clarifying questions** to enable the skill. 4. Set the **Max questions per turn** (1–10, default 3). This caps how many questions the agent can ask in a single invocation. ⚠️ This skill is model-dependent. Smaller or less capable models may not reliably choose to call the tool, even when clarification would be helpful. For best results, use a capable model as your agent LLM. ![AnythingLLM Agent Settings Menu Icon Location](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Fanythingllm-setup%2Fagent-configuration%2Fsettings-menu-icon-location.png&w=3840&q=100) ![AnythingLLM Agent Settings Menu](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fagent-surveys%2Fmodal.png&w=3840&q=100) How It Works[](https://docs.anythingllm.com/features/agent-surveys#how-it-works) --------------------------------------------------------------------------------- Once enabled, the agent has access to a tool that lets it present questions to the user. The agent autonomously decides when clarification is needed based on the prompt it received. ### Question Types[](https://docs.anythingllm.com/features/agent-surveys#question-types) | Type | Description | | --- | --- | | **Text input** | Free-form text, URL, number, date, or email fields | | **Single-select choice** | Pick one option from a list | | **Multi-select choice** | Pick one or more options from a list | Choice questions can optionally include an "Other" field so the user can provide a custom answer that wasn't in the original list. ### User Experience[](https://docs.anythingllm.com/features/agent-surveys#user-experience) When the agent asks questions, the user sees an interactive card in the chat: * **Single question** — A simple form with the question and input or choices. * **Multiple questions** — A paginated card the user can navigate through. * **Skip** — Users can skip individual questions or dismiss the entire survey if they'd rather let the agent proceed on its own. After the user responds (or skips/times out), the agent receives the answers and continues its task with that information in context. ![AnythingLLM Multi-Choice Survey](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fagent-surveys%2Fmulti-choice.png&w=3840&q=100) ### Persistence[](https://docs.anythingllm.com/features/agent-surveys#persistence) Completed surveys are saved alongside the chat message. When you reload a conversation, surveys render as read-only cards showing the questions and answers that were provided. The Q&A transcript is also injected into prompt history, so the agent retains awareness of what was discussed in future turns. ![AnythingLLM Saved Survey](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fagent-surveys%2Fsaved-survey.png&w=3840&q=100) Last updated on June 24, 2026 [AI Agents](https://docs.anythingllm.com/features/ai-agents "AI Agents") [Private Browser Tool](https://docs.anythingllm.com/features/browser-tool "Private Browser Tool") --- # Authenticated Scraping ~ AnythingLLM [🚀 AnythingLLM v1.15.0 is live! Update now →](https://anythingllm.com/download) Feature Overview Private Browser Tool 🖥️ **Desktop Only Feature (v1.8.3+)** The Authenticated Scraping tool is exclusively available in the AnythingLLM Desktop application! Authenticated Scraping ====================== ⚠️ **Security Note** All credentials and session data are stored locally on your machine. AnythingLLM never transmits or stores your login information outside of your local machine. The Authenticated Scraping tool enables you to access and scrape gated online content from websites or services that require authentication but hold critical contexual content you might want to use in your workflows, such as your personal LinkedIn feed or internal company portals that you have access to. Your LLM can now access these websites and scrape and view content just like you would in a regular browser! ![](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fbrowser-tool%2Fbrowser-tool.png&w=1080&q=100) Key Features[](https://docs.anythingllm.com/features/browser-tool#key-features) -------------------------------------------------------------------------------- * **Secure Session Storage**: Credentials are stored locally using isolated browser sessions * **Session Persistence**: Login sessions persist between app restarts until explicitly cleared or the authentication expires for the associated service * **Isolated Environment**: Separate from your actual web browser * **Full User Control**: Clear stored data or sessions at any time with a single click Using the Authenticated Scraping Tool[](https://docs.anythingllm.com/features/browser-tool#using-the-authenticated-scraping-tool) ---------------------------------------------------------------------------------------------------------------------------------- ### Accessing the Tool[](https://docs.anythingllm.com/features/browser-tool#accessing-the-tool) 1. Open AnythingLLM Desktop 2. Navigate to Settings > Tools > Browser Tool 3. Click "Open Private Browser" to launch the isolated browser window ![](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fbrowser-tool%2Fmanager.png&w=2048&q=100) ### Authentication Process[](https://docs.anythingllm.com/features/browser-tool#authentication-process) 1. Log into your desired service (e.g., LinkedIn, Gmail) through the Authenticated Scraping tool. 2. Your session will persist until you explicitly clear the browser data or the authentication expires for the associated service. 3. AnythingLLM can now access authenticated content from these services when scraping or via agentic workflow execution. 4. The returned content will be text only. No images, videos, or other media will be returned. 💡 **Heads up!** The Authenticated Scraping tool is not a magic bullet. It is a tool that allows you to access authenticated content from websites that require authentication. It cannot currently **interact** with the content of the page you are accessing (eg: browser automation, RPA, etc). ### Managing Browser Data[](https://docs.anythingllm.com/features/browser-tool#managing-browser-data) * **Clearing Data**: Use the "Clear Browser Data" button to remove all stored credentials and sessions * **When should you clear the browser data?**: * When switching between different service accounts * If you encounter authentication issues when the LLM tries to access the site you want to scrape Common Use Cases[](https://docs.anythingllm.com/features/browser-tool#common-use-cases) ---------------------------------------------------------------------------------------- ⚠️ **Warning!** Some web services may detect and restrict automated access, even though this tool functions as a standard browser. Use this feature responsibly and at your own discretion, as certain services may suspend or block accounts that they perceive as engaging in automated activity. * Scraping your personal linkedin profile or feed. * Accessing internal company documentation that is behind a login or SSO portal. * Collecting or accessing data from paid or authenticated web service you have access to normally. Troubleshooting[](https://docs.anythingllm.com/features/browser-tool#troubleshooting) -------------------------------------------------------------------------------------- If you encounter issues: 1. Clear the browser data and try again 2. Ensure you're fully logged into the service by opening the private browser and navigating to the site you want the LLM to access. Some services have very short lived sessions, those services may require you to log in again after a certain amount of time or might be a bad use-case for this tool. You can always re-authenticate with the service by opening the private browser and navigating to the site you want the LLM to access and logging in again to refresh the session. Last updated on June 24, 2026 [Agent Surveys](https://docs.anythingllm.com/features/agent-surveys "Agent Surveys") [API Access](https://docs.anythingllm.com/features/api "API Access") --- # Chat Modes ~ AnythingLLM [🚀 AnythingLLM v1.15.0 is live! Update now →](https://anythingllm.com/download) Feature Overview Chat Modes Chat Modes in AnythingLLM ========================= AnythingLLM offers multiple ways to chat with your documents. Let's understand what each does and how to get the best results for your use case and expectations. Available Chat Modes[](https://docs.anythingllm.com/features/chat-modes#available-chat-modes) ---------------------------------------------------------------------------------------------- **Agent Mode** (recommended): 💡 All versions later than v1.11.1 have agent mode available as an option for workspaces, enabled by default for new workspaces. You can change the chat mode in the workspace's settings by clicking the "Gear" icon under the "Chat Settings" tab. * Will automatically use available agent-skills, tools, and MCPs to answer your questions * This feature is fully dependent on the capabilities of your LLM provider and model to call tools natively. * Native tool calling is enabled by default for providers that support it, giving you an always-on agent experience. If a provider's native tool calling misbehaves, you can fall back to prompt-based tool calling with the [`PROVIDER_DISABLE_NATIVE_TOOL_CALLING`](https://docs.anythingllm.com/configuration#native-tool-calling-for-llm-providers) environment variable. _if you see the "@" symbol in your prompt input you will need to use `@agent` to start an agentic chat session. If it is not there, you are using agentic chat mode._ **Chat Mode**: * Uses both your documents and the AI's general knowledge * More conversational and flexible * Good for brainstorming and exploring topics **Query Mode**: * Only uses information from your uploaded documents * Will tell you if it can't find relevant information * Best for when you need accurate, document-based answers and nothing else Common Questions[](https://docs.anythingllm.com/features/chat-modes#common-questions) -------------------------------------------------------------------------------------- ### "It keeps saying 'No relevant information found' in Query mode"[](https://docs.anythingllm.com/features/chat-modes#it-keeps-saying-no-relevant-information-found-in-query-mode) This usually means one of three things: 1. The information might be in your document but worded differently 2. The similarity settings might be too strict 3. The document might be too large and split in a way that makes finding information difficult **Quick fixes to try:** 1. Go to workspace settings → Vector Database Settings 2. Change "Document similarity threshold" to "No restriction" 3. Try asking your question using words that match how it's written in your document 💡 Instead of asking "How do I start the app?", try using terms from your document like "How do I initialize the application?" ### "When should I use Query mode vs Chat mode?"[](https://docs.anythingllm.com/features/chat-modes#when-should-i-use-query-mode-vs-chat-mode) Use **Query mode** when: * You need factual answers from your documents * You're working with technical documentation * You want to prevent made-up information Use **Chat mode** when: * You want more conversational responses * You need additional context or examples * You're brainstorming ideas ### "Why does it work better with some documents than others?"[](https://docs.anythingllm.com/features/chat-modes#why-does-it-work-better-with-some-documents-than-others) Documents are processed in chunks, and each chunk is analyzed separately. This means: * Large documents might need more specific questions * Technical documents work better with technical questions Tips for Better Results[](https://docs.anythingllm.com/features/chat-modes#tips-for-better-results) ---------------------------------------------------------------------------------------------------- 1. **Start with Query mode** and "No restriction" similarity if you're not finding information 2. **Use specific terms** from your documents in your questions 3. **Switch to Chat mode** if you need more context or explanation 4. **Try rephrasing your question** if you're not getting good results ⚠️ If you're still not getting good results, check your workspace settings and try adjusting the "Document similarity threshold" between No restriction, Low (≥ .25), Medium (≥ .50), or High (≥ .75) to find what works best for your documents. [Chat Logs](https://docs.anythingllm.com/features/chat-logs "Chat Logs") [Embedded Chat Widgets](https://docs.anythingllm.com/features/chat-widgets "Embedded Chat Widgets") --- # API Access & Keys ~ AnythingLLM [🚀 AnythingLLM v1.15.0 is live! Update now →](https://anythingllm.com/download) Feature Overview API Access ![AnythingLLM](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fapi%2Fheader-image.png&w=3840&q=100) API Access & Keys ================= You can find the API documentation for available endpoints on your instance at `/api/docs` API keys are managed by accounts with the correct access level. However, anyone with the API key can use the AnythingLLM API, so do not share or publish this key anywhere. AnythingLLM supports a full developer API that you can use to manage, update, embed, and even chat with your workspaces. You can create and delete API keys on the fly if you are allowed permission to do so. ![AnythingLLM](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fapi%2Fapi-keys.png&w=3840&q=100) Last updated on June 24, 2026 [Private Browser Tool](https://docs.anythingllm.com/features/browser-tool "Private Browser Tool") [Appearance Customization](https://docs.anythingllm.com/features/customization "Appearance Customization") --- # Appearance Customization ~ AnythingLLM [🚀 AnythingLLM v1.15.0 is live! Update now →](https://anythingllm.com/download) Feature Overview Appearance Customization ![AnythingLLM Appearance Customization](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fcustomization%2Fheader-image.png&w=3840&q=100) Appearance Customization ======================== ️💡 **DOCKER VERSION ONLY!** These settings are only available in the Docker version of AnythingLLM AnythingLLM allows you to customize the look and feel of your instance to match your brand and identity. ![Appearance Settings Page](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fcustomization%2Fappearance-settings-page.png&w=3840&q=100) Overview of all the appearance settings available in AnythingLLM. Custom Logo[](https://docs.anythingllm.com/features/customization#custom-logo) ------------------------------------------------------------------------------- ![Custom Logo](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fcustomization%2Fcustom-logo.png&w=3840&q=100) You can replace the AnythingLLM branded logo that appears on the login page and throughout the app with your own brand's logo. In this example, we have used a green square image for demonstration purposes. Custom Welcome Messages[](https://docs.anythingllm.com/features/customization#custom-welcome-messages) ------------------------------------------------------------------------------------------------------- ![Custom Welcome Messages](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fcustomization%2Fcustom-welcome-messages.png&w=3840&q=100) By default, when you first log in to AnythingLLM and you have not yet selected a workspace, you will be shown the default messages explaining AnythingLLM. Using the system messages inputs, you can simulate both system and user response messages. Take this opportunity to tell users what specific workspaces are for - or just say hello! Custom Footer Links and Icons[](https://docs.anythingllm.com/features/customization#custom-footer-links-and-icons) ------------------------------------------------------------------------------------------------------------------- ![Custom Footer Links and Icons](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fcustomization%2Fcustom-footer-links-and-icons.png&w=3840&q=100) The footer icons can be replaced with custom links and icons to provide quick access to relevant resources or web pages. Last updated on June 24, 2026 [API Access](https://docs.anythingllm.com/features/api "API Access") [Chat Logs](https://docs.anythingllm.com/features/chat-logs "Chat Logs") --- # What is AnythingLLM ~ AnythingLLM [🚀 AnythingLLM v1.15.0 is live! Update now →](https://anythingllm.com/download) ![AnythingLLM](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Fgetting-started%2Fintroduction%2Fheader-image.png&w=3840&q=100) What is AnythingLLM? ==================== AnythingLLM is the easiest to use, **all-in-one** AI application that can do RAG, AI Agents, and much more with no code or infrastructure headaches. AnythingLLM is built by [Mintplex Labs, Inc (opens in a new tab)](https://github.com/Mintplex-Labs) - founded by [Timothy Carambat (opens in a new tab)](https://twitter.com/tcarambat) and went through [YCombinator Summer 2022 (opens in a new tab)](https://www.ycombinator.com/companies/mintplex-labs) . AnythingLLM is **not a one-person project**. The Mintplex Labs team also includes: * Sean Hatfield (Engineer) * Marcello Fitton (Engineer) * Tiff Tang (Designer) * Our community of volunteer contributors * [You? (opens in a new tab)](https://www.ycombinator.com/companies/mintplex-labs/jobs) Why use AnythingLLM? ==================== You want a **zero-setup**, **private**, and all-in-one AI application for local LLMs, RAG, and AI Agents all in one place without painful developer-required set up. [Learn more about AnythingLLM Desktop →](https://docs.anythingllm.com/installation-desktop/overview) _or_ You need a **fully-customizable**, **private**, and all-in-one AI app for your _business or organization_ that is basically a full ChatGPT with permissioning but with any LLM, embedding model, or vector database. [Learn more about AnythingLLM for Docker →](https://docs.anythingllm.com/installation-docker/local-docker) If either of these things excite you - you will love watching the video below. [AnythingLLM Roadmap](https://docs.anythingllm.com/roadmap "AnythingLLM Roadmap") [All Features](https://docs.anythingllm.com/features/all-features "All Features") --- # Workspace Chat Logs ~ AnythingLLM [🚀 AnythingLLM v1.15.0 is live! Update now →](https://anythingllm.com/download) Feature Overview Chat Logs ![AnythingLLM Workspace Chat Logs](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fchat-logs%2Fheader-image.png&w=3840&q=100) Workspace Chat Logs =================== AnythingLLM supports exporting chats as: * **CSV** * **JSON** * **JSON (Alpaca)** * **JSONL (OpenAI fine-tune)** Just click export at the top of the screen once at least 10 chat logs are available! Provided you have the correct account permissions, you can view the chat logs per workspace and per user of your AnythingLLM instance. ![AnythingLLM Workspace Chat Logs](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fchat-logs%2Fworkspace-chat.png&w=3840&q=100) [Appearance Customization](https://docs.anythingllm.com/features/customization "Appearance Customization") [Chat Modes](https://docs.anythingllm.com/features/chat-modes "Chat Modes") --- # Embedded Chat Widgets ~ AnythingLLM [🚀 AnythingLLM v1.15.0 is live! Update now →](https://anythingllm.com/download) Feature Overview Embedded Chat Widgets ![AnythingLLM Embedded Chat Widgets](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fchat-widgets%2Fheader-image.png&w=3840&q=100) Embedded Chat Widgets ===================== ️💡 **DOCKER VERSION ONLY!** These settings are only available in the Docker version of AnythingLLM ![Embedded Chat Widget](https://docs.anythingllm.com/_next/image?url=%2Fimages%2Ffeatures%2Fchat-widgets%2Fchat-widget.png&w=3840&q=100) AnythingLLM allows you to create embedded chat widgets that can be easily integrated into any website using a simple `