Generate, test, retrieve and validate standalone audio and video assets through a self-hosted ComfyUI server, local or remote: MiniMax Music 3 for music beds and jingles, Stable Audio 3 for SFX, MiniMax H3 text-, image- and reference-to-video with multi-frame guidance. Use for "generate BGM / a jingle / sound effects", "make a short video from this image", "MiniMax", "Stable Audio", "the ComfyUI queue is busy", "free the model before switching", or operating existing ComfyUI workflows over LAN or a private network while scheduling GPU work safely (one job at a time, /history is the only proof of completion). The Unity side is unity-audio.
