Avatar Shorts — LongCat-Video-Avatar 1.5

Upload a portrait image and an audio narration to generate a talking-head video with lip-sync. Audio is chunked at silence boundaries (~3.7 s per clip); each clip uses the previous clip's last frame for visual continuity. Practical limit ~5 minutes on ZeroGPU xlarge.

API: client.predict(image, audio, prompt, api_name='/generate')