ddcat-ai/open-ai-canvas

What is it?

Key features

• 01Infinite canvas with nodes, connections, minimap, undo/redo and read-only sharing.
• 02Structured storyboards, character and scene management, and a 3D directing stage.
• 03Text, image, video and audio generation with reference frames and camera control.
• 04Async task queue with progress, logs, cancel and retry.
• 05Timeline editing with subtitle transcription and server-side video export.
• 06Admin tools for model channels, usage quotas and object storage (Alibaba Cloud OSS, Tencent COS).

Use cases

• Drafting a short drama from script to storyboard and generated shots on one canvas.
• Organizing character, scene and style references for consistent AI video production.
• Self-hosting a shared creative workspace for a small trusted team.

Quick start

1. 1

   Clone the repository

   git clone https://github.com/ddcat-ai/open-ai-canvas.git
2. 2

   Create local data and cache directories

   mkdir -p .local/project-workbench-debug .local/cache/go-build .local/cache/go-mod
3. 3

   Start the Go backend in its own terminal (run from the backend directory)

   cd backend && CANVAS_BACKEND_ADDR=127.0.0.1:8080 CANVAS_BACKEND_DATA_DIR=../.local/project-workbench-debug GOCACHE=../.local/cache/go-build GOMODCACHE=../.local/cache/go-mod go run ./cmd/server
4. 4

   Install frontend dependencies in web/, then run bun run dev and open http://localhost:3000

   cd web && bun install --frozen-lockfile

Things to know

• !The project is still under rapid development, so data structures and external interfaces may change.
• !It is suited to personal, local or trusted environments and should not be exposed as a public multi-user service without security configuration.
• !Running it requires Go 1.25 and Bun, and you must configure your own model channels with API keys for generation.

In collections

Topics

FAQ

What is open-ai-canvas?

Open-source AI film and short-drama creation workspace with an infinite canvas for storyboarding, multimodal generation and asset management. Open AI Canvas (影策) is a self-hosted workspace for turning a story into video. It uses a free-form canvas where scripts, characters, scenes, style boards and reference materials are arranged as connected nodes, with structured storyboards and a 3D directing stage for planning shots.

Is open-ai-canvas free to use? What license does it have?

open-ai-canvas is open source under the MIT license. Check the license terms before using it in commercial products.

Is open-ai-canvas still maintained?

Actively maintained. The last commit to open-ai-canvas was on Nov 6, 2026.

What can I use open-ai-canvas for?

Drafting a short drama from script to storyboard and generated shots on one canvas. Organizing character, scene and style references for consistent AI video production. Self-hosting a shared creative workspace for a small trusted team.

How do I install open-ai-canvas?

Clone the repository: git clone https://github.com/ddcat-ai/open-ai-canvas.git Create local data and cache directories: mkdir -p .local/project-workbench-debug .local/cache/go-build .local/cache/go-mod Start the Go backend in its own terminal (run from the backend directory): cd backend && CANVAS_BACKEND_ADDR=127.0.0.1:8080 CANVAS_BACKEND_DATA_DIR=../.local/project-workbench-debug GOCACHE=../.local/cache/go-build GOMODCACHE=../.local/cache/go-mod go run ./cmd/server Install frontend dependencies in web/, then run bun run dev and open http://localhost:3000: cd web && bun install --frozen-lockfile

Where can I get open-ai-canvas?

The source code and installation instructions are on GitHub: https://github.com/ddcat-ai/open-ai-canvas