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

TaylorSeerBufferState

`TaylorSeerBufferState`

class max.pipelines.diffusion.TaylorSeerBufferState(factor_0, factor_1, factor_2, last_compute_step=None)

source

Bases: object

Per-request mutable TaylorSeer state using Buffer objects.

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

Parameters:

factor_0 (Buffer)
factor_1 (Buffer)
factor_2 (Buffer)
last_compute_step (int | None)

`factor_0`

factor_0: Buffer

source

Cached 0th-order factor (function value).

`factor_1`

factor_1: Buffer

source

Cached 1st-order factor (first derivative approximation).

`factor_2`

factor_2: Buffer

source

Cached 2nd-order factor (second derivative approximation).

`last_compute_step`

last_compute_step: int | None = None

source

Step index of the last full transformer computation.

TaylorSeerBufferState​

factor_0​

factor_1​

factor_2​

last_compute_step​

`TaylorSeerBufferState`

`factor_0`

`factor_1`

`factor_2`

`last_compute_step`