Livestream Update: @MiniMax_AI H3: A New Open-Weight Video Model, Live in ComfyUI MiniMax H3 is an open-weight, general-purpose multimodal video generation model that works across text, images, video, and audio. In ComfyUI, you can use H3 for text-to-video, image-to-video, first- and last-frame generation, and reference-driven creation. H3 jointly generates the visuals and synchronized stereo audio, including dialogue, sound effects, ambience, and music, rather than adding audio afterward. The open-weight H3 checkpoints support clips up to 15 seconds at 768p. MiniMax’s hosted H3 model also supports generation at up to 2K resolution. During the stream, we’ll test the model live and discuss how H3 brings multiple generation tasks into one architecture, how its high-compression video representation improves efficiency, and what developers should know when setting it up locally through ComfyUI. What we'll cover: →MiniMax H3: open weights, native stereo audio, up to 2K/15s clips →Live demo →How H3 got small enough to run on consumer hardware →ComfyUI workflow templates for H3 →Local workflow: use the H3 Context-IR API ( generate locally at 768p, and use the H3 Regenerate-2K API