The CausVid tool uses a full-sequence diffusion model to teach an autoregressive system to swiftly create videos. Users can generate clips and modify them on the fly with new prompts, turn a photo into moving scenes, and extend videos.

The CausVid generative AI tool uses a diffusion model to teach an autoregressive (frame-by-frame) system to rapidly produce stable, high-resolution videos.

  • A_A@lemmy.worldEnglish
    2·
    2 hours ago

    Arguably about 100 fold improvement over "diffusion models” like OpenAl’s SORA and Google’s VEO 2.
    Such major improvements are to be expected in my opinion in the future of artificial intelligence.