anodyine/vllm-omni
1
0 Fork 0
Code Issues 4 Pull Requests 2 Actions Packages Projects Releases Wiki Activity

diffusion: enable VAE patch parallel for SD3.5 (#1428)

Browse Source 
Signed-off-by: dongbo910220 <1275604947@qq.com>
This commit is contained in:
dongbo910220 authored and GitHub committed 2026-02-27 10:28:30 +08:00
1 parent 3b0c21f94d
commit f6ed1bc4e7
2 files changed
+3 -2
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

docs/user_guide/diffusion/parallelism_acceleration.md
+2 -2
View File
@@ -32,7 +32,7 @@ The following table shows which models are currently supported by parallelism me
| **Qwen-Image-Edit-2509** | `Qwen/Qwen-Image-Edit-2509` | ✅ | ✅ | ✅ | ✅ | ❌ |
| **Qwen-Image-Layered** | `Qwen/Qwen-Image-Layered` | ✅ | ✅ | ✅ | ✅ | ❌ |
| **Z-Image** | `Tongyi-MAI/Z-Image-Turbo` | ✅ | ✅ | ❌ | ✅ (TP=2 only) | ✅ |
| **Stable-Diffusion3.5** | `stabilityai/stable-diffusion-3.5` | ❌ | ❌ | ❌ | ✅ | ❌ |
| **Stable-Diffusion3.5** | `stabilityai/stable-diffusion-3.5` | ❌ | ❌ | ❌ | ✅ | ✅ |
| **FLUX.2-klein** | `black-forest-labs/FLUX.2-klein-4B` | ❌ | ❌ | ❌ | ✅ | ❌ |
| **FLUX.1-dev** | `black-forest-labs/FLUX.1-dev` | ❌ | ❌ | ✅ | ✅ | ❌ |
@@ -85,7 +85,7 @@ outputs = omni.generate(
VAE patch parallelism distributes the VAE decode workload across multiple ranks by splitting the latent spatially. It is configured via `DiffusionParallelConfig.vae_patch_parallel_size` and can be combined with other parallelism methods (e.g., TP).
!!! note "Enablement and feature gate"
- VAE patch parallelism is currently **enabled only for validated pipelines** (currently: `Tongyi-MAI/Z-Image-Turbo`).
- VAE patch parallelism is currently **enabled only for validated pipelines** (currently: `Tongyi-MAI/Z-Image-Turbo`, `stabilityai/stable-diffusion-3.5`).
- If `vae_patch_parallel_size > 1` is set for a validated pipeline, vLLM-Omni will automatically enable `vae_use_tiling` as a safety gate. (We use `vae_use_tiling` because it indicates the VAE supports diffusers tiling parameters like `tile_latent_min_size` and `tile_overlap_factor`.)
#### Offline Inference