Synaplan
A self-hosted AI platform connects document search, agents and customer chat, with local models through Ollama or cloud providers.

Synaplan is an open-source AI platform for teams that want assistants to answer questions from company files and handle customer conversations. It brings internal knowledge and customer support into a shared system on infrastructure you control.
- Document knowledge: Search documents, email and files with RAG, so assistants can use company material when answering questions.
- Agents with approvals: Give assistants tools for business tasks, require human approval before write actions, and review an audit log of their steps.
- Customer channels: Use the same knowledge base and agent rules across a website chatbot, WhatsApp Business, Outlook and email; support staff can take over widget conversations.
- Model choice and costs: Route tasks to Ollama, OpenAI, Anthropic, Gemini, Groq or xAI, and see the cost of each answer.
- Connected work: Link Microsoft 365, Nextcloud and calendars, extend agents through plugins and MCP, and save tasks to run on demand or on a schedule.
Synaplan runs on your own hardware or servers through Docker or Kubernetes. It uses the Apache License 2.0, and the hosted service in Germany runs the same open-source software. iPhone and Android apps can connect to your own server or the hosted service.
With local models through Ollama, the platform can operate offline or in an air-gapped environment without external AI API calls. Cloud model use sends requests to the selected provider and requires its credentials; channels such as WhatsApp and Microsoft 365 depend on external services.