AnythingLLM
Open-source AI workspace for private document chat, agents, meetings, and local or managed deployment.
About AnythingLLM
AnythingLLM is an open-source AI workspace for document knowledge, research, agent workflows, and local meeting assistance. The desktop app can run on macOS, Windows, and Linux without an account or cloud API key, while organizations can self-host a multi-user deployment on their own infrastructure. Documents can become part of an agent knowledge base without leaving the device, and the meeting assistant can create transcripts, summaries, action items, and key decisions locally without adding a bot to the call. Teams can also configure models, permissions, custom agent workflows, white-labeling, and developer integrations. Deployment spans a free MIT-licensed desktop or self-hosted edition, private hosted Basic and Pro instances, and an Enterprise option for on-premise deployment. Hosted Basic costs $50 monthly and requires an LLM API key; Pro costs $99 monthly and includes priority resources and a 72-hour support SLA. Enterprise pricing is available by consultation. Local and self-hosted use gives teams more control over data placement, but hardware capacity, deployment maintenance, and any external model-provider charges remain part of the operating decision.
Best for
Best use cases
A document collection needs to become a searchable local knowledge base without the files leaving the device.
A meeting needs local transcription, summaries, action items, and key decisions without a bot joining the call.
An internal AI workspace needs private hosting, user isolation, administrative controls, and custom agent workflows.
Compare before choosing
Choose the desktop edition when local privacy and a no-account setup matter more than centralized team administration.
Choose self-hosting when infrastructure control and multi-user access justify deployment and maintenance work.
Choose a hosted plan when a managed private instance is preferable, while accounting for the required model API key on Basic.
Features
Private document research
Build a local knowledge base from documents, then ask questions and run research without sending the document collection to a cloud service.
On-device meeting follow-up
Transcribe meetings locally and produce summaries, action items, key decisions, and full transcripts without a bot joining the call.
Custom team AI workspace
Deploy a multi-user instance, configure models and permissions, and create custom agent workflows for an internal team.
Pros
- The free desktop and self-hosted editions offer local control without requiring an account or cloud API key.
- Document knowledge and meeting transcription can run on the device, reducing the need to send source material to cloud services.
- Desktop, self-hosted, managed cloud, and enterprise on-premise options cover individual and multi-user workflows.
Cons
- Self-hosted deployments require infrastructure setup, maintenance, and enough hardware capacity for the selected local models.
- The $50 monthly Basic hosted plan requires a separate LLM API key, which can add model-provider usage costs.
- Advanced team controls and managed support are tied to hosted or enterprise deployment choices rather than the basic desktop workflow.
Pricing Plans
Desktop / Self-hosted
- MIT-licensed open-source software
- Runs locally with no account or cloud API key required
- Supports a desktop installation or a multi-user self-hosted deployment
Basic
- Private hosted instance
- Custom subdomain
- RAG and agents ready to use
- Bring your own LLM API key
Pro
- Private hosted instance with priority resources
- RAG and agents ready to use
- Designed for teams that rely on AI daily
- 72-hour support SLA
Enterprise
- On-premise deployment
- Custom support SLA
- Custom integrations
- SSO and RBAC
* Pricing information is for reference only. Please refer to the latest announcement on AnythingLLM's official website.