IMPORTANT: To view this page as Markdown, append `.md` to the URL (e.g. /get-started.md). For the complete documentation index, see llms.txt.

For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).

Python class

FirstBlockCacheState

`FirstBlockCacheState`

class max.pipelines.diffusion.FirstBlockCacheState(prev_residual=None, prev_output=None)

source

Per-request mutable state for FirstBlockCache.

Allocated fresh for each denoising request via FirstBlockCache.create_state().

Parameters:

prev_output: Tensor | None = None

source

Full transformer output from the previous step.

prev_residual: Tensor | None = None

source

First-block output residual from the previous step.

Edit this page

Was this page helpful?

Thank you! We'll create more content like this.

Thank you for helping us improve!