Local Minimax H3 Music Video Generation Workflow Shared
Old-Age6220 · reddit · 2026-08-20
A user shared a workflow for creating music videos using local AI tools. The core setup involves Minimax H3 for music/audio generation and ComfyUI for video generation. The process utilizes ComfyUI's default text-to-video templates, followed by upscaling to produce the final video.
More from Multimodal
- Test: Generating a 30-sec car commercial with WAN3.0 in one go — alifcoder · 2026-08-20
- A reusable papercut stop-motion JSON prompt template for recreating movie scenes with MiniMax H3 — techhalla · 2026-08-20
- QuerySplat: Decoupling Geometry and Appearance in 3DGS Prediction — rsasaki0109 · 2026-08-20
- A painter noticed AI models keep generating the same names and motifs — creativity or convergence? — MenegattiArt · 2026-08-20
- Running WAN 2.2 on a laptop to make 5-second idle loop animations — and asking for prompt tips — J_Lezter · 2026-08-20
- GTA: San Andreas Reimagined in Claymation Style — ShaKodemon · 2026-08-20