This repo introduces LongCat-Video, a 13.6B-parameter model that handles text-to-video, image-to-video, and video continuation within a single framework. It uses block sparse attention and a coarse-to-fine strategy to produce minutes-long 720p/30fps videos without quality drift. The project also includes an audio-driven avatar extension with Whisper-based lip sync and distillation-accelerated inference.

• A single 13.6B-parameter model handles text-to-video, image-to-video, and video continuation, producing minutes-long 720p/30fps clips without quality drift.
• Avatar 1.5 (May 2026) switched from wav2vec2 to Whisper-Large-v3 for better lip sync, improved long-video stability, and added stylized domains like anime and animal content.
• Step-distilled inference cuts video generation down to just eight steps, with INT8 quantization also supported.

Google has released Veo 3.1 Lite, a budget-friendly video generation model that allows developers to create high-volume applications at over 50% less cost than Veo 3.1 Fast, while maintaining the same speed. The model supports Text-to-Video and Image-to-Video features, and is accessible through the Gemini API and Google AI Studio.

• Veo 3.1 Lite costs over 50% less than Veo 3.1 Fast while matching its speed
• Supports Text-to-Video and Image-to-Video in 16:9 or 9:16 up to 1080p, with 4, 6, or 8 second durations priced by length
• Google is also cutting prices on Veo 3.1 Fast alongside this launch
• Available now via Gemini API and Google AI Studio for paid tier users