ChatGLM Releases CogVideoX-5B-I2V Model for Image-to-Video Generation; Qwen2-VL-2B Supports Video Processing

The release of the CogVideoX-5B-I2V model by the ChatGLM team marks a significant advancement in open-source AI. This new model allows users to generate videos from images, using an image as a background along with prompts. The model is now available on GitHub and can be run on free-tier Colab and Gradio space. Additionally, the Qwen2-VL-2B model developed by the Alibaba Qwen team now supports both video and image processing, thanks to contributions from the AI community. Users are advised to duplicate the space with a L4s to avoid long waiting queues.