Goku
ByteDance + HKU streaming video foundation model — generates digital humans
A ByteDance + HKU streaming video generation foundation model that can generate digital humans directly.
AI video generation, digital humans, and video restoration.
Key features
- Rectified flow Transformer for joint image-and-video generation
- Supports text-to-video, image-to-video, and text-to-image tasks
- 84.85 VBench text-to-video score; top scores on GenEval and DPG-Bench
- Fine-grained data curation and enhanced video-image token interaction
Good for
- Text-to-video and image-to-video generation
- Foundation model research on rectified flow
Stack & tags
video-generationtext-to-videoimage-to-videotext-to-imagerectified-flowopen-weightsfoundation-model
Similar open-source projects
Hyperframes
★ 55,468Write HTML. Render video. Built for agents — HeyGen's open-source video framework
Hypit
★ 18,649Clone any viral video with AI agents — swap the face, the words, the B-roll, ship 100 variants