chenchenghao/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

[diffusion] feat: support cache-dit integration (#14234)

Browse Source 
Co-authored-by: shuxiguo <shuxiguo@meituan.com>
Co-authored-by: DefTruth <qiustudent_r@163.com>
Co-authored-by: Mick <mickjagger19@icloud.com>
This commit is contained in:
blahblah
2025-12-06 00:52:22 +08:00
committed by GitHub
co-authored by shuxiguo DefTruth Mick
parent 889b46ea50
commit 66984a8b3d
9 changed files with 826 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files
python/sglang/multimodal_gen/docs/environment_variables.md
+19
View File
@@ -0,0 +1,19 @@
## Cache-DiT Acceleration
These variables configure cache-dit caching acceleration for Diffusion Transformer (DiT) models.
See [cache-dit documentation](cache_dit.md) for details.
| Environment Variable | Default | Description |
|-------------------------------------|---------|------------------------------------------|
| `SGLANG_CACHE_DIT_ENABLED` | false | Enable Cache-DiT acceleration |
| `SGLANG_CACHE_DIT_FN` | 1 | First N blocks to always compute |
| `SGLANG_CACHE_DIT_BN` | 0 | Last N blocks to always compute |
| `SGLANG_CACHE_DIT_WARMUP` | 4 | Warmup steps before caching |
| `SGLANG_CACHE_DIT_RDT` | 0.24 | Residual difference threshold |
| `SGLANG_CACHE_DIT_MC` | 3 | Max continuous cached steps |
| `SGLANG_CACHE_DIT_TAYLORSEER` | false | Enable TaylorSeer calibrator |
| `SGLANG_CACHE_DIT_TS_ORDER` | 1 | TaylorSeer order (1 or 2) |
| `SGLANG_CACHE_DIT_SCM_PRESET` | none | SCM preset (none/slow/medium/fast/ultra) |
| `SGLANG_CACHE_DIT_SCM_POLICY` | dynamic | SCM caching policy |
| `SGLANG_CACHE_DIT_SCM_COMPUTE_BINS` | not set | Custom SCM compute bins |
| `SGLANG_CACHE_DIT_SCM_CACHE_BINS` | not set | Custom SCM cache bins |