seam
Give it a start and end photo and it generates the AI video transition between them (smoke, water bend, whip pan) on your own GPU, no cloud render needed.
View jdnichollsc/seam on GitHub- GitHub stars
- 26
- Forks
- —
- Language
- Python
Project details
- Repository
- jdnichollsc/seam
- Licence
- MIT
- Primary language
- Python
- Catalog status
- Available
- Catalog check
- 2026-11-13
Repository description
Two shots in, one transition out. Open-source AI video transitions (smoke, water bend, whip pan) from a first and last image, rendered on your own GPU with open-weights LTX models via ComfyUI + the Comfy SDK. Web app, CLI and a AI agent plugin.
Explore by topic
More tools in Video & animation
- scroll-studio
okohedeki · AGPL-3.0
Scroll Studio turns a single YAML file into a scrolling, AI-generated website: locally generated video, hand-drawn artwork, live 3D zooms, data stories, and map…
- skills
scenario-labs · MIT
A skill pack that lets any AI agent generate ready-to-use images, video, audio and 3D assets while keeping characters and brand style consistent, and even drive…
- dioramas
blendi-remade · MIT
An open-source framework that turns a text prompt into a 3D hero object (via AI image + image-to-3D generation) and drops it into a cinematic, interactive landing page,…
- comfyui-h3-facerefine
carasibana · MIT
A ComfyUI add-on that automatically detects, crops, and re-renders small faces in MiniMax H3-generated video so they come out sharp instead of blurry.