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

[diffusion] feat: support torch compile for diffusers backend (#19673)

Browse Source
This commit is contained in:
DefTruth
2026-03-04 14:08:45 +08:00
committed by GitHub
parent c6850ac30c
commit e9b5706545
5 changed files with 74 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
docs/diffusion/api/cli.md
View File

@@ -286,6 +286,8 @@ SGLang diffusion supports a **diffusers backend** that allows you to run any dif
| `--vae-slicing` | flag | Enable VAE slicing for lower memory usage (decodes slice-by-slice). |
| `--dit-precision` | `fp16`, `bf16`, `fp32` | Precision for the diffusion transformer. |
| `--vae-precision` | `fp16`, `bf16`, `fp32` | Precision for the VAE. |
| `--enable-torch-compile` | flag | Enable `torch.compile` for diffusers pipelines. |
| `--cache-dit-config` | `{PATH}` | Path to a Cache-DiT YAML/JSON config file for accelerating diffusers pipelines with Cache-DiT. |
### Example: Running Ovis-Image-7B