Updated Minimax Music Video Workflow
retained text
Workflow here: [Dantesworks/MV\_Maker: Makes music videos for ComfyUI/H3 Minimax](https://github.com/Dantesworks/MV_Maker/tree/main) This is a music video comprised of multiple clips. Each clip is sent the exact part of the song it corresponds to hence lip syncing is easy and automatic (if prompted to do so). Each clip has a "theme" - for example, cafe, garden etc. You can specify the themes at the start and assign a theme to each clip. Decide the timing of the clips too. What is missing is face refiner. You also need a way to assemble all the clips at the end. I have an all in one face refiner and clips assembly to run on your selected clips at the end but its slop coded and not fit for release. You can manually edit the prompts for each clip, but I recommend Codex or Claude to do it for you instead. Have fun! I know I did :)
document fetched 2026-10-30 00:32 · canonical https://reddit.com/r/StableDiffusion/comments/1xiajrd/updated_minimax_music_video_workflow