JD.com Enters Open-Source AI Video with JoyAI-Echo

Chinese e-commerce company JD.com has released JoyAI-Echo, a new open-source model for generating video from text. The release marks the company's entry into the competitive field of open-source generative video, adding another major corporate player to the ecosystem.

Unlike many models that produce short, single clips, JoyAI-Echo is designed for creating "long-form, multi-shot" videos. This allows it to generate a sequence of related scenes that can form a more coherent narrative. Crucially, the model also generates synchronized audio to accompany the video, a feature still emerging in many open video tools.

The model is based on the LTX-Video research framework, which focuses on generating temporally consistent and longer video sequences. The full model, code, and weights are available on its Hugging Face repository under an Apache 2.0 license, which permits commercial use.

JoyAI-Echo’s release highlights the growing trend of moving beyond simple clip generation toward more practical, narrative-driven video creation. Its focus on multi-shot storytelling and integrated audio pushes the capabilities of what's available in open-source AI, offering a new tool for creators and researchers exploring long-form generative content.