Skip to content

Commit 98cba68

Browse files
committed
add magcache to caching api listing
1 parent 48f39c2 commit 98cba68

2 files changed

Lines changed: 7 additions & 3 deletions

File tree

docs/source/en/api/cache.md

Lines changed: 7 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -35,8 +35,14 @@ Cache methods speedup diffusion transformers by storing and reusing intermediate
3535

3636
[[autodoc]] apply_first_block_cache
3737

38-
### TaylorSeerCacheConfig
38+
## TaylorSeerCacheConfig
3939

4040
[[autodoc]] TaylorSeerCacheConfig
4141

4242
[[autodoc]] apply_taylorseer_cache
43+
44+
## MagCacheConfig
45+
46+
[[autodoc]] MagCacheConfig
47+
48+
[[autodoc]] apply_mag_cache

docs/source/en/optimization/cache.md

Lines changed: 0 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -118,8 +118,6 @@ pipe.transformer.enable_cache(config)
118118

119119
MagCache relies on **Magnitude Ratios** (`mag_ratios`), which describe this decay curve. These ratios are specific to the model checkpoint and scheduler.
120120

121-
### Usage
122-
123121
To use MagCache, you typically follow a two-step process: **Calibration** and **Inference**.
124122

125123
1. **Calibration**: Run inference once with `calibrate=True`. The hook will measure the residual magnitudes and print the calculated ratios to the console.

0 commit comments

Comments
 (0)