Open-source ComfyUI fork processes up to 9 faces per pass, cutting MiniMax H3 video face work nearly 5x
This_Temporary_8537 · reddit · 2026-09-08
A Reddit user released ComfyUI-H3-FaceRefine-Accelerated, a fork that speeds up multi-face cleanup in MiniMax H3 videos.
- Packs multiple tracked faces into one atlas and runs them through H3 in a single pass (up to 9 faces), instead of serial per-face runs
- Splits long videos into chunks and carries latent context across joins for smoother transitions
- On a 4090, a 15-second clip with four faces dropped from 16.5 minutes to 3.5 minutes
- Trade-off: 4 faces in a 768 atlas means 384px per face, so single-face path is still best for close-ups
- Backward compatible; accelerated nodes are optional
Related event: Open-Source Fork Speeds Up MiniMax H3 Face Refinement by Nearly 5x(2 posts)→
More from Multimodal
- Fish Audio ships open-weight Drama 3 TTS; creator shows turning old posts into narrated videos — alex_verem · 2026-10-27
- Redditor crafts photorealistic 'morning bus ride' short film entirely in ComfyUI — no3us · 2026-10-27
- New blur/anti-blur slider LoRA for MOSS lets you dial background sharpness with weights — NoMouse9610 · 2026-10-27
- PixGR: Pixel-space generative decoding for fast feed-forward 3D reconstruction — kwangmoo_yi · 2026-10-27
- Krea-2 Turbo + ComfyUI for product ads: prompts and results shared — comfy-king-999 · 2026-10-27
- High-res tiled Marigold V2 depth maps arrive in ComfyUI with See Through model — AntonObukhov1 · 2026-10-27