ZhipuThe technical team today released andOpen Sourcelatest versionVideo Model CogVideoX v1.5, compared to the original model, CogVideoX v1.5 will include 5/10 second, 768P, 16-frame video generation capability, I2V model support for any size scale, dramatically improving the quality of graph-generated video and complex semantic understanding.
Officially, CogVideoX v1.5 will also be synchronized to the "ClearVideo" platform, and combined with the newly launched CogSound sound model, the "new ClearVideo" will have the following features:
- Quality Improvement: significant increase in ability in quality, aesthetic presentation, motion rationalization, and semantic comprehension of complex cue words in graphic born videos.
- Ultra HD resolution: support for generating 10s, 4K, 60 fpsUltra HD video.
- Variable Scale: SupportArbitrary ratioThe playback is very simple and can be adapted to different playback scenarios.
- Multi-channel output: the same command/image canGenerate 4 videos at once.
- AI video with sound effects: New ClearShadow can generateSound effects to match the graphics.
Attach the open source address below:
Code:
- https://github.com/thudm/cogvideo
Model:
- https://huggingface.co/THUDM/CogVideoX1.5-5B-SAT
statement:The content is collected from various media platforms such as public websites. If the included content infringes on your rights, please contact us by email and we will deal with it as soon as possible.