Excaliself vs Excalidraw+
Excaliself and Excalidraw+ both build a team whiteboard on the Excalidraw canvas. The core decision is deployment model: Excalidraw+ is a managed cloud service with built-in AI features; Excaliself is self-hosted and MIT licensed.
| Feature | Excaliself | Excalidraw+ |
|---|
| License | MIT | Proprietary |
| Hosting | Self-hosted (Docker + Postgres) | Managed cloud |
| Team space | Yes | Yes |
| Live collaboration | Yes | Yes |
| AI features | Via MCP (client-side, any MCP client) | Built-in AI tools |
| Pricing | Free (server costs only) | Paid subscription |
| Data location | Your Postgres database | Excalidraw's servers |
| Import from Excalidraw+ | Yes (beta) | n/a |
Excaliself is the better choice when the team needs drawings stored on their own infrastructure, wants no per-seat licensing cost, or needs to connect AI clients through MCP without sending data to a third party. MIT licensing means teams can fork, modify, or extend it. The beta import path from Excalidraw+ makes it a direct migration target for teams leaving the managed service.
Excalidraw+ is worth considering when the team wants AI features built directly into the canvas (Excaliself has no built-in AI, only MCP access from connected clients), when managed hosting is preferable over self-managed Postgres and Docker infrastructure, or when the team has no data-residency requirement and prioritizes minimal setup over data ownership.