Collected sources and patterns will appear here. Add from search, explore, or the patterns library.
Tensor<VideoFrames> -> Tensor<DiscreteTokens>
Quantize raw video frames into discrete integer tokens.
Problem it solves
Autoregressive generative modeling of visual states requires representing physical video frames as sequences of categorical values.
Consumes
Emits
The real projects this mechanism was found in. Attribution is the point — this is how the best teams actually do it.