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

[FP8 Quantization] Add FP8 quantization support for Flux transformer (#1640)

Browse Source 
Signed-off-by: zhou zhuoxin <zhouzhuoxin1508@outlook.com>
This commit is contained in:
zhou zhuoxin authored and GitHub committed 2026-03-05 11:08:34 +08:00
1 parent 5493e75877
commit 58eb1c68db
3 files changed
+64 -11
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

docs/user_guide/diffusion/quantization/fp8.md
+1
View File
@@ -62,6 +62,7 @@ The available `ignored_layers` names depend on the model architecture (e.g., `to
|-------|-----------|---------------|------------------|
| Z-Image | `Tongyi-MAI/Z-Image-Turbo` | All layers | None |
| Qwen-Image | `Qwen/Qwen-Image`, `Qwen/Qwen-Image-2512` | Skip sensitive layers | `img_mlp` |
| Flux | `black-forest-labs/FLUX.1-dev` | All layers | None |
## Combining with Other Features