Open-Source Afflatus Turns ComfyUI Workflows into One-Click Forms for Image and Video Generation
Basic_Aside_7547 · reddit · 2026-10-11
Developer bardak971 released Afflatus, a free open-source UI layer on top of ComfyUI aimed at generating images and videos without touching the node graph.
Key features:
- Import your own ComfyUI workflows as simple forms, with built-in forms for common models
- All outputs land in a library with projects and folders; any render can be reopened with its settings
- One-click tools: image-to-video, upscaling, face fixing, inpainting, background removal
- A timeline for assembling video clips with takes, trims, and music
ComfyUI still does all the work; Afflatus just replaces graph-dragging with a single clean screen — the snowboard film in the demo was made with it. First release, Linux/Mac builds untested, feedback welcome.
More from Multimodal
- Authors First raises $10M+ to adapt Genghis novel into AI-made six-episode series — lmoroney · 2026-10-11
- User has Claude generate a cinematic music video for their own track — Pleasant_Salt6810 · 2026-10-11
- Someone turned bad AI consciousness takes into a Suno-generated bingo song — dioscuri · 2026-10-11
- Building a physically realistic grand piano simulator down to hammer felt hysteresis with Grok — doodlestein · 2026-10-11
- AI Trailer Wins Inaugural $2.5M Future Vision XPRIZE, Heading for $10M Feature Development — lmoroney · 2026-10-11
- Extending Qwen Image 2.1 Turbo's magic 8-step sigmas to 12-14 steps — GTManiK · 2026-10-11