[diffusion] Enable SANA-Video breakable CUDA graphs (#35729)

This commit is contained in:
Xiaoyu Zhang
2026-08-22 12:57:09 +08:00
committed by GitHub
parent 83e9ece672
commit 96bfd2476c
7 changed files with 104 additions and 7 deletions
+13
View File
@@ -234,6 +234,19 @@ sglang serve --model-path meituan-longcat/LongCat-Image \
--port 30010
```
SANA-Video supports the same path for fixed-resolution serving. Its default
text stage emits a fixed 300-token prompt shape, so the runtime reuses one
graph across prompt lengths without padding it to a generic text bucket:
```bash Command
sglang serve \
--model-path Efficient-Large-Model/SANA-Video_2B_480p_diffusers \
--enable-breakable-cuda-graph \
--warmup-resolutions 832x480 \
--enable-torch-compile false \
--port 30010
```
### Component Residency
Use `--component-residency COMPONENT=MODE` to assign one runtime residency mode to each native pipeline component: