MiniMax AI is launching its new open-weight video generation model, H3, on August 7th at 10 AM PT. The model supports text, image, video, and audio inputs and can generate up to 2K resolution clips lasting 15 seconds. The launch event, hosted in collaboration with ComfyUI, will feature a live demonstration, insights into H3's efficient architecture, and ready-to-use ComfyUI workflow templates. AI
IMPACT This release of an open-weight video model could accelerate research and development in AI-driven video generation.
RANK_REASON Frontier-lab model release with system card. [lever_c_demoted from frontier_release: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
