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

[diffusion] Default Hunyuan VAE to tiled decode (#36012)

Browse Source
This commit is contained in:
Xiaoyu Zhang
2026-08-24 13:18:06 +08:00
committed by GitHub
parent f98b60de80
commit 344613c159
3 changed files with 34 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/docs/sglang-diffusion/api/cli.mdx
+9
View File
@@ -212,6 +212,15 @@ vae_sp: true
enable_torch_compile: false
```
HunyuanVideo and FastHunyuan use tiled VAE decode by default so multi-GPU runs
distribute VAE tiles instead of selecting spatial-shard decode. At
HunyuanVideo's supported 960×544×77 shape, spatial-shard decode can consume
99.8 GiB per rank before requesting another 49.61 GiB causal mask. At
FastHunyuan's default 1280×720×125 shape, the mask alone would require
197.75 GiB. You can still override the policy with
`--vae-config.parallel-decode-mode`, but `spatial` and `spatial_shard` should
only be used for smaller validated shapes.
## Generate
`sglang generate` runs a single generation job and exits when the job finishes.