MiniMax H3 Day-0 Support in ComfyUI: Open Weights, Native Audio, and 2K Video

MiniMax released its H3 open-weights video model with native stereo audio and 2K output on Monday, and ComfyUI added day-zero support for the model the same morning. The company said H3 is its third-generation video model and the first released with open weights. The model accepts text, images, video, or audio inputs and generates video clips up to 15 seconds long with stereo sound produced in the same pass. It supports text-to-video, image-to-video, first-and-last-frame control, and reference-to-video workflows that carry a subject, motion, or voice through a clip. MiniMax said H3's multimodal context understanding collapses five separate tasks into one model by taking images, audio, and video together and resolving them against a prompt that explains how they relate. The company said motion transfer allows a reference video to supply movement while the subject and style come from elsewhere, and combined with in-place editing enables iterating on a shot.