From $20 per month🇺🇸
Desktop app for downloading and running open language models locally via llama.cpp and MLX, with an agent and a command line.
AnythingLLM AI features and pricing (October 2026)
Open-source AI app from Mintplex Labs for chat with documents and agents, as a desktop app, self-hosted or in the cloud.
FreemiumFree plan
from $50 per month
Pricing details
Desktop app and self-hosting with Docker are free (MIT license). Cloud per the pricing page: Basic $50, Pro $99 per month, Enterprise with on-premises, SSO and RBAC by quote, as of September 2026.
What is AnythingLLM?
AnythingLLM is an open-source AI application for chatting with documents and running agents, available as a desktop app, a self-hosted multi-user server or a hosted cloud instance, built by Mintplex Labs.
The desktop app runs on macOS, Windows and Linux without dependencies, and an Android app runs models on the device. According to the README, the Docker version supports several users on one server with permissions. Features include a document knowledge base, agents with custom skills, background jobs, a meeting assistant and web search. The code is MIT licensed; the website reports more than 66,000 GitHub stars and names companies such as Merck, Nokia and Red Hat among its users.
AnythingLLM AI features and use cases
- Answers questions from internal documentsTeams load manuals, policies or project files into a workspace and ask questions in chat. The app retrieves relevant passages and answers with a local or cloud model, which saves searching through folders.
- Summarizes meetingsThe meeting assistant transcribes and summarizes conversations, according to the website. Participants get notes without sending audio to a separate transcription service when local models are used.
- Automates recurring tasks with agentsAgents with custom skills and background jobs run tasks such as research or document processing. Small teams can automate routine work without a separate agent platform.
What sets AnythingLLM apart
- Same open-source code base as desktop app, multi-user Docker server and hosted cloud, so teams can start small and move to a server later.
- SSO and role-based access control are listed for the Enterprise tier, which also offers on-premises deployment.
Where AnythingLLM is strongest
United States Available internationally
Desktop and self-hosted software that runs in any country. Mintplex Labs lists the United States as its location.
Questions to ask before you buy AnythingLLM
Decide between desktop, self-hosted and cloud before rollout, since only Enterprise lists SSO and RBAC. For the cloud plans, ask Mintplex Labs where instances are hosted and for a data processing agreement, which we did not find published. Document which model processes which data when cloud models are connected.
AnythingLLM reviews
No reviews yet. Be the first to share how your team uses it.
Every review shows its verification level. We read each one before it goes live. How we check reviews
Questions about AnythingLLM
Is AnythingLLM free?
The desktop app and self-hosting with Docker are free under the MIT license. Hosted cloud plans start at $50 per month as of September 2026.
Can several employees use one installation?
Yes. According to the README, the Docker version supports multiple users with permissions; the desktop app does not.
Does AnythingLLM support SSO?
SSO and role-based access control are listed for the Enterprise tier, which is priced by quote.
AnythingLLM update historyno updates yet
No changes recorded for AnythingLLM yet.
Changes the vendor made and confirmed user reports, each checked against the vendor's own pages. We review every tool monthly.
Alternatives to AnythingLLM
Tools in the same category (Desktop apps) with the most overlapping markets.
From $149 per year🇺🇸
Desktop and web workspace for local and hosted language models, with document knowledge stacks, model comparison and agents.
From $20 per month🇺🇸
Model runner for open language models on your own machine, with a local API and optional cloud models.
Free🇸🇬
Open-source desktop app from Menlo Research that runs open models offline and optionally connects to cloud models.