Ask a question below.
Published answers will appear here.
This isn't just a video model, it's a streaming, autoregressive diffusion transformer. It sequentially predicts frame by frame, tackling the challenge of accumulating errors in generation. Plus, it's interactive, allowing text prompts to guide the stream! https://x.com/nathanbenaich/status/2077015304954147131/video/1
This isn't just a video model, it's a streaming, autoregressive diffusion transformer. It sequentially predicts frame by frame, tackling the challenge of accumulating errors in generation. Plus, it's interactive, allowing text prompts to guide the stream! https://x.com/nathanbenaich/status/2077015304954147131/video/1
Ask a question below.
Published answers will appear here.