Live Avatar Alibaba

Live Avatar Alibaba is an algorithm-system co-designed framework that enables real-time, streaming, and infinite-length interactive avatar video generation. It is designed for users who need to engage in natural, face-to-face conversations via microphone and camera, receiving immediate visual feedback as the avatar responds in real-time. The framework is powered by a 14-billion-parameter diffusion model that achieves 45 FPS on 5 H800 GPUs with 4-step sampling, supporting Block-wise Autoregressive processing for streaming video generation up to 10,000+ seconds long.

The key capabilities of Live Avatar Alibaba include its ability to generate frames faster than playback speed, support unbounded, continuous streaming expansion based on preceding frames, and mitigate inference-mode drift, distribution drift, and error accumulation. It achieves this by employing Distribution Matching Distillation, designing Timestep-forcing Pipeline Parallelism, and applying System-Level Optimizations. Live Avatar Alibaba also integrates with Qwen3-Omni to enable fully interactive dialogue agents.

Live Avatar Alibaba is most valuable for users who need to create interactive avatar videos for various applications, such as virtual assistants, chatbots, or virtual reality experiences. These users can benefit from the framework's real-time, streaming, and infinite-length capabilities, as well as its ability to mitigate common issues with long-horizon generation. By using Live Avatar Alibaba, users can create more realistic and engaging interactive avatar videos that can be used in a variety of contexts.