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

[diffusion] feat: automatically infer comfy fp8 activation scaling (#36060)

Browse Source
This commit is contained in:
Mick
2026-08-23 18:43:33 +08:00
committed by GitHub
parent 44db041700
commit dd15fb57b5
5 changed files with 53 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/docs/sglang-diffusion/quantization.mdx
+3 -1
View File
@@ -51,7 +51,9 @@ repo contains multiple candidate checkpoints, pass
`--transformer-weights-path` explicitly.
MiniMax-H3 is a verified example for Comfy safetensors with per-layer metadata,
including `pruned_fp8_scaled` and serialized ConvRot INT8. Pass one selected
including `pruned_fp8_scaled` and serialized ConvRot INT8. Other Comfy FP8
exports are also auto-detected: the presence of an input scale selects static
activation scaling, while its absence selects dynamic scaling. Pass one selected
FL2VA or Ref2VA DiT file by local path, `owner/repo/path/file.safetensors`, or
direct Hugging Face file URL; do not combine it with `--quantization`. Its GGUF
usage is documented in