For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /max/get-started.md).
Python class
FirstBlockCacheState
FirstBlockCacheState
class max.pipelines.diffusion.FirstBlockCacheState(prev_residual=None, prev_output=None)
Bases: object
Per-request mutable state for FirstBlockCache.
Allocated fresh for each denoising request via
FirstBlockCache.create_state().
prev_output
Full transformer output from the previous step.
prev_residual
First-block output residual from the previous step.
Was this page helpful?
Thank you! We'll create more content like this.
Thank you for helping us improve!