Open-Source AI Canvas Connects Image-to-Video Workflow
GeeseHomard · reddit · 2026-07-18
Someone shared an open-source AI canvas project that manages multi-step image-to-video workflows within a single reusable graph, reducing the friction of switching between tools. Its workflow includes:
- Using the built-in 3D Director for camera blocking, allowing setup of characters, poses, camera angles, and simple geometry
- Feeding the composition into an image generation node
- Reusing the image result as a reference, first frame, or last frame for video generation
- Keeping all intermediate results on the same infinite canvas
- Supporting saving graphs in-browser or exporting as JSON
Supported image models include GPT Image, Nano Banana, and Seedream; video models include Kling, Seedance, and Veo. The author notes a key limitation: the app is MIT-licensed and self-hostable, but models don't run locally; generation and media transfer rely on the paid SJolt API, with API keys and graphs stored in the browser. The post concludes by asking if this built-in 3D blocking will replace parts of the Blender/ComfyUI workflow for image-to-video creators, or if the reliance on cloud APIs will be a dealbreaker.
Related event: Open-Source AI Canvas Integrates Image-to-Video Workflows(2 posts)→
More from coding & agent
- Harness bug masked Gemini 4 Argon: rerun with Antigravity beats Astra, nears Opus 5.5 — vitaliychiley · 2026-10-27
- Design Is Becoming Game Dev: Build Your Own Tools With AI as the Executor — hewarsaber · 2026-10-27
- Dev turns DJI Mavic into local AI security drone built with Opus 5.5 — alexcovo_eth · 2026-10-27
- Microsoft Research: ForkPilot cuts long-horizon agent compute by up to 59% — EliasEskin · 2026-10-27
- Cursor launches Agent SDK: agents as code with Markdown prompts and TypeScript tools — soleio · 2026-10-27
- Vibecheck, the 'usable formal methods' hackathon, sells out two floors in SF — spikedoanz · 2026-10-27