The company wants video models to stream frame by frame instead of rendering after a prompt.
Google's updated video model reads ten seconds of context, pins start and end frames, and renders drafts in…
LTX-2.5 brings open-weights video generation to ComfyUI with multishot and audio support.
MiniMax's H3 model generates 2K video with native stereo audio from mixed text, image and sound prompts.
xAI's latest video generation model doubles rendering speed while adding parallel agent support and searchable project libraries.