synaplan
Omnichannel AI platform connecting your knowledge to web, WhatsApp, email, Outlook and cloud storage.
Deploy AI assistants across web, WhatsApp, email, Outlook, and cloud storage with per-answer cost tracking.
Synaplan is an open-source AI platform that connects your documents to multiple communication channels including web widgets, WhatsApp, and email. It suits support and operations teams needing unified AI responses backed by local or cloud models with strict cost tracking.
What it is
The platform indexes your files to answer queries across web apps, mobile clients, Outlook, Nextcloud, and messaging channels while displaying the model and price for every single answer. It features a planner that routes requests to expensive models only when necessary and pauses critical actions like sending mail for human approval.
Who it's for
It suits organizations wanting to deploy identical AI knowledge bases across customer support channels without paying per-user SaaS fees. It does not suit teams looking for a simple, single-container chat window like Open WebUI, as Synaplan requires managing multiple containers and services.
Key features
Omnichannel Deployment
Answer questions on your website, WhatsApp, email, Outlook, Nextcloud, and ownCloud with the same knowledge base.
Per-Answer Cost Tracking
Every answer displays its specific model and price to keep your usage expenses transparent.
Hybrid Model Routing
Mix local Ollama models with cloud providers like OpenAI, Anthropic, Google, and Mistral per task.
Human-in-the-Loop Actions
Configure actions like sending mail or creating tickets to wait for explicit manual approval.
Up and running in 3 steps
Our honest verdict
What's great
- Connects local and cloud AI models side-by-side with automatic cost tracking per response
- Embeds directly into WhatsApp, Outlook, email, and Nextcloud alongside the web widget
- Maintains strict control over data with self-hosting options and no training on user inputs
- Includes a planning step that delegates tasks to expensive models only when required
Where it falls short
- Requires Docker with Compose v2 and at least 8 GB of RAM to run locally
- First-time setup involves multiple containers and a download phase lasting up to 15 minutes
- More complex to deploy than single-container chat alternatives if you only need a basic web interface
Pricing & license
Self-hosting the software is free under the Apache-2.0 license, though you must pay for your own server hardware, database upkeep, and cloud AI provider API keys. A hosted version is also available with paid plans.
Open remote AI-training roles
879 open remote AI-training roles, updated November 15, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.
Alternatives
These reviewed AI tools do the same job instantly — from our scored directory.
Use it, or have us set it up
FAQ
Can I run models locally without sending data to the cloud?
Yes, you can connect local models via Ollama and run them side-by-side with cloud providers.
Which communication channels are supported out of the box?
Synaplan supports web widgets, WhatsApp, email, Outlook, Nextcloud, ownCloud, and OpenCloud.
Is there a hosted version available?
Yes, a hosted version is available with paid plans or invitation codes, while self-hosting is free under the Apache-2.0 license.
