Skip to the content
Know your tech before you buy it. Makers: get listed from $49/mo

AnythingLLM

anythingllm.com

  • Free plan
  • Windows
  • macOS
  • Linux
  • Web
  • Android
  • API
  • Self-hosted
8.6 Club score Visit anythingllm.com

What AnythingLLM is

AnythingLLM is an AI assistant with desktop, mobile, self-hosted, and cloud options. Its desktop app runs on macOS, Windows, and Linux, and the free Desktop plan requires no account or API key and has no token limits. You can turn documents into an agent knowledge base that stays on the device. Desktop tools include web search and scraping, file-system capabilities, model selection, background jobs, custom agent skills, and a meeting assistant that records, transcribes, summarizes, and indexes meetings on the computer without a bot joining or cloud processing. The mobile app runs small models on-device and offers web search, document chat, file generation, scheduled jobs, and memory. Supported model providers range from local options such as Ollama and LM Studio to OpenAI, Anthropic, and Google Gemini Pro. Cloud instances support multiple users, tenant isolation, admin controls, and white-labeling. Self-hosting is available, and the maker says instance data remains on the user's infrastructure. Optional telemetry can be disabled. Cloud Basic costs $50 a month; Pro costs $99 a month.

Who it suits

It suits individuals who want an on-device assistant and teams that need cloud administration or self-hosted deployment. Mobile users can also chat with documents and run small models on-device.

Good to know, and what to watch

  • Free desktop plan has no account or API key requirement.
  • Documents can form an on-device agent knowledge base.
  • Meeting assistant works without a bot or cloud processing.
  • Supports desktop, mobile, self-hosted, and cloud options.
  • Optional telemetry can be disabled.
  • Windows documentation targets Windows 11.
  • Cloud Basic costs $50 a month.
  • Cloud Pro costs $99 a month.

Basic

$50/mo

Pro

$99/mo

Enterprise

—

Desktop

Free

Free plan

Getting started

  1. Open https://anythingllm.com/
  2. Choose Desktop, Mobile, self-hosted, or a cloud instance
  3. Install the desktop build for macOS, Windows, or Linux, or use the Android app
  4. Select an LLM provider such as Ollama, OpenAI, or another supported integration
  5. Upload documents to build an agent knowledge base
  6. Choose a plan if using cloud administration

Limits to check

The Desktop plan is free with no account, API key, or token limit.

The Club's review

Overview

AnythingLLM brings document chat, AI agents, and meeting summaries together in desktop, mobile, cloud, and self-hosted versions. It suits people who want to work with private documents on their own devices, as well as teams that need cloud administration. Its strongest case is the choice between local use and managed deployments; Windows users should check the supported version guidance before installing.

The core software is released under the MIT License, and the desktop app can run without an account, an API key, or a token limit. That makes it a practical starting point for individual use, though cloud plans cost more as collaboration and support needs grow. For other options, see Retrieval-Augmented Generation Tools and AI chatbots and assistants.

Key features

Documents, models, and agents

AnythingLLM turns documents into an agent knowledge base while keeping the data on the device in its local setup. Desktop features include file-system access, web scraping and search, model selection, background jobs, and custom agent skills. This mix suits users who want to build a document assistant around their own workflows; the Developer API extends the product where built-in functions do not cover a need.

Model choices include local runtimes such as Ollama, LM Studio, Local AI, KobaldCPP, and oMLX, plus hosted providers including OpenAI, Azure OpenAI, Anthropic, Cohere, Google Gemini Pro, Hugging Face, Together AI, OpenRouter, Perplexity AI, Mistral API, and Groq. That breadth helps users keep a local workflow or bring an existing provider key, though using hosted models may involve separate provider costs.

Meeting Assistant

The desktop Meeting Assistant records, transcribes, summarizes, and indexes meetings on the computer. It works without a bot joining the call or cloud processing, a useful fit for people who want meeting notes to stay local.

Mobile and cloud

AnythingLLM Mobile runs small models on-device and includes web search, document chat, file generation, scheduled jobs, and memory. Cloud instances add multi-user workspaces, tenant isolation, fine-grained administration, and white-labeling, which are aimed at organizations managing access across users.

For self-hosted deployments, Mintplex Labs says it does not host or access instance documents, chats, workspace settings, or embeddings; those remain on the user's infrastructure. Telemetry is optional and anonymous, excludes document content and chat logs, and can be disabled in application settings.

Pricing

AnythingLLM uses a freemium model. Desktop is free at no cost, with no account, API key, or token limit. That is the clearest choice for individuals who can run the assistant locally; a token limit-free app does not remove any charges a user may incur from a separately chosen hosted model provider.

PlanPriceWhat it includesBest suited to
Self-hostedCustom pricingDocker-based, open-source deploymentTeams that want to operate their own deployment
Basic$50 a month, billed monthlyPrivate instance, custom subdomain, RAG, agents, and bring-your-own LLM API keySmall deployments that need a private cloud instance
Pro$99 a month, billed monthlyPrivate instance, RAG, agents, priority resources, and a 72-hour support SLATeams that need priority resources and a defined support response
EnterpriseCustom pricingOn-premise deployment, custom SLA and integrations, SSO, and RBACOrganizations requiring tailored administration or identity controls

Basic costs less than Pro but does not include the priority resources or 72-hour support SLA described for Pro. Neither cloud plan's user-seat allowance or usage quotas are stated, so teams should weigh the monthly price against their expected scale. Self-hosted and Enterprise use custom pricing, making them harder to compare on cost alone.

Platforms

AnythingLLM supports Android, Linux, macOS, self-hosted deployments, web, Windows, and an API. Native desktop builds cover macOS on Apple Silicon and Intel, Windows on x64 and ARM, and Linux. The Windows application targets Windows 11 and is intended for a Windows Home user account; other Windows versions may not work. Check that guidance before making it the basis of a Windows rollout.

Who it's for

AnythingLLM is a strong fit for individuals who want a local document assistant without creating an account or supplying an API key, and for users who value local meeting transcription and summaries. Developers can draw on the API and broad model integrations, while teams can choose cloud administration or keep deployments on their own infrastructure.

It is less compelling for buyers who need a clear per-seat cloud price or published usage limits before committing: those terms are not provided for the Basic and Pro plans. Windows users outside the stated Windows 11 target should also consider whether another supported platform better fits their setup.

Pros and cons

  • Pros: The free desktop app has no account, API key, or token limit, lowering the barrier to a local assistant.
  • Pros: Document knowledge and meeting processing can stay on the computer, useful when keeping work local matters.
  • Pros: Local and hosted model integrations give users flexibility in how they connect to language models.
  • Pros: Mobile, self-hosted, and multi-user cloud options cover different deployment needs.
  • Cons: Basic starts at $50 a month and Pro at $99 a month, with no stated seat counts or usage quotas to help teams forecast total cost.
  • Cons: Windows guidance targets Windows 11 and a Home user account, limiting confidence for other Windows setups.
  • Cons: Self-hosted and Enterprise pricing is custom, so cost comparison requires a quote.

Alternatives

Choose Flowise if a flow-and-assistant builder with a published free tier suits better: its free plan allows 2 Flows & Assistants, 100 Predictions / month, and 5MB Storage. Choose Google Cloud Agent Evaluation if you need API-based agent evaluation with usage-based computation metrics rather than a general local assistant.

Haystack is an open-source Python framework for teams building around a framework. Haystack Enterprise Platform offers a Studio tier with one workspace, one user, 100 pipeline hours, and 50 files up to 10MB each. PrivateGPT, txtai, I3K RAG Enterprise, and RAGFlow are other options in the document and RAG space.

Verdict

Choose AnythingLLM if you want a flexible assistant for your documents, especially when local data handling, on-device meetings, or a free desktop start matters. Its range of deployment choices is the main reason to pick it. Look elsewhere if your decision depends on predictable cloud costs per seat, or if your Windows environment falls outside the stated target.

The verdict

AnythingLLM offers several deployment choices, with on-device document handling and meeting tools alongside cloud administration. Check the Windows version guidance and cloud plan costs against your needs.

Questions people ask

Is AnythingLLM free?

Yes. The Desktop plan is listed as free and includes no account, no API key, and no token limits. A free plan is also available in the pricing model.

Which platforms does AnythingLLM support?

Platforms listed are Android, API, Linux, macOS, self-hosted, web, and Windows. Native desktop builds cover macOS Apple Silicon or Intel, Windows x64 or ARM, and Linux.

Can AnythingLLM be self-hosted?

Yes. The Self-hosted plan uses Docker and is an open-source deployment. The maker says instance documents, chats, workspace settings, and embeddings remain on the user's infrastructure.

Does AnythingLLM support integrations?

Yes. Listed providers include Ollama, LM Studio, OpenAI, Azure OpenAI, Anthropic, Cohere, Google Gemini Pro, Hugging Face, Together AI, OpenRouter, Perplexity AI, Mistral API, Groq, and generic OpenAI.

What the maker publishes

Cloud administration
Cloud instances support multiple users with tenant isolation, fine-grained admin controls, and white-labeling. source
Desktop features
The desktop product lists file-system capabilities, web scraping and search, dynamic model selection, background jobs, custom agent skills, and Meeting Assistant. source
Developer API
The maker describes a full Developer API for functionality not already built into the product. source
Document knowledge
The product turns documents into an agent knowledge base while keeping the data on the device. source
File uploads
Yes source
Free plan
Yes source
Image generation
Yes source
App integrations
Yes source
License
The AnythingLLM core is provided under the MIT License. source
LLM integrations
Supported providers include Ollama, LM Studio, Local AI, KobaldCPP, oMLX, OpenAI, Azure OpenAI, Anthropic, Cohere, Google Gemini Pro, Hugging Face, Together AI, OpenRouter, Perplexity AI, Mistral API, Groq, and generic OpenAI. source
Meeting assistant
Meeting Assistant records, transcribes, summarizes, and indexes meetings on the computer without a bot joining the call or cloud processing. source
Mobile
AnythingLLM Mobile runs small models on-device and includes web search, document chat, file generation, scheduled jobs, and memory. source
Mobile apps
Yes source
Self-hosted data
For self-hosted deployments, Mintplex Labs says it does not host, store, or access instance documents, chats, workspace settings, or embeddings, which remain on the user's infrastructure. source
Paid from
50 source
Support
The Pro cloud plan includes a 72-hour support SLA, while Enterprise offers a custom SLA. source
Supported desktop systems
Native builds are provided for macOS with Apple Silicon or Intel, Windows with x64 or ARM, and Linux. source
Team workspace
Yes source
Telemetry
Telemetry is optional, anonymous, excludes document content and chat logs, and can be disabled in the application settings. source
Web search
Yes source
What it does
AnythingLLM is a private AI assistant that runs entirely on the user's computer, with no account, API key, or token limit required. source
Windows limitation
The Windows documentation says the application targets Windows 11, is intended for a Windows Home user account, and that other Windows versions may not work. source

Checked against the maker's pages on 28 Sept 2026.