CogVideoX

Zhipu AI announced today that it will open source CogVideoX, a video generation model with the same origin as "Qingying". According to reports, the CogVideoX open source model includes multiple models of different sizes. Currently, CogVideoX-2B will be open source. It requires 18GB of video memory for inference at FP-16 precision and 40GB of video memory for fine-tuning. This means that a single 4090 graphics card can perform inference, and a single A6000 graphics card can complete fine-tuning. The upper limit of the prompt word of CogVideoX-2B is 226...

According to Sina Technology, Zhang Peng, CEO of Zhipu AI, was interviewed today and talked about the impact of generative AI-based video technology on the film and television industry. He said that since the emergence of OpenAI's Sora, the topic has been discussed. Abroad, this matter has caused a great impact, including the Hollywood strike. Zhang Peng revealed that he once discussed this matter with teachers at the Beijing Film Academy, and everyone came to a unanimous view: from the perspective of technological development, everyone believes that the development of AI is "a good thing" and "a very important direction", which is...

Zhipu AI announced today that it has upgraded its video generation model and officially launched a new generation product - CogVideoX. The CogVideoX model is now available on Zhipu Qingyan's PC, mobile application, and mini program. All C-end users can experience the AI text-generated video and image-generated video services for free through Zhipu Qingyan's AI video generation function "Qingying". It is reported that the core technical features of CogVideoX are as follows: In order to solve the problem of content coherence, Zhipu AI independently developed...