Lip Sync Test with MiniMax H3: Local and Cloud Workflow
HorkBorkington · reddit · 2026-08-28
A user shared a lip sync test using MiniMax H3, partially generated locally on a 5070ti and sped up with cloud GPU. The workflow involved a custom script to chop audio, prompting with reference images/video, and reassembling in Premiere.
More from Multimodal
- Halcyon early-fusion image models match late-fusion with 275x less pretraining compute, open-sourced — davidwromero · 2026-10-16
- Running Wan2.1 14B FP16 on a 12GB laptop GPU: VRAM swap stress test finishes in 4m11s — RYuu-1120 · 2026-10-16
- Striking Midjourney generations: black holes and planetary aesthetics — ciguleva · 2026-10-16
- Fable 5.5 may drop today as Andrew Curran hints at early release — kimmonismus · 2026-10-16
- Kling 4.0 Adds Timeline Editing, Endorsed by 'Seoul Spring' Director Kim Seong-su — xiaohu · 2026-10-16
- One prompt, no reference images: Argon builds 3D Gaussian splats from geometry equations in NumPy — fofrAI · 2026-10-16