MiniMax Launches Omni-Modal Video Model Generating 15-Second 2K Clips With Native Stereo Audio
AI video model releases
MiniMax Launches Omni-Modal Video Model Generating 15-Second 2K Clips With Native Stereo Audio
MiniMax launched its new video model on August 1, 2026, available via platform API and the consumer Hailuo AI app.
The model generates 2K resolution video clips lasting 15 seconds with native stereo audio built in, rather than added in post-production.
It unifies text-to-video, image-to-video, subject referencing, and motion editing into a single pretraining paradigm, replacing pipelines that previously needed separate expert models for each task.
Users can mix sources for control, such as taking camera movement from one reference and character voice from another, all directed through natural language prompts.
Target use cases include advertising, e-commerce, product design, and film pre-visualization, with the model accessible only through cloud infrastructure rather than local hardware.
cho.sh