AX1Y2JP
/

Krea-2-Turbo-INT8-ConvRot

Model card Files Files and versions

AX1Y2JP commited on Jul 2

Commit

8bef355

·

verified ·

1 Parent(s): 5ae9bba

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -12,7 +12,7 @@ library_name: diffusers
 inference: false
 ---
-⚠️ComfyUI's native int8 support only supports tensorwise, so this model won't work. Please continue to use int8-fast.
 A Krea 2 model quantized to INT8 with ConvRot using a conservative quantization policy.
 <img src="ComparisonA.webp" width="1000">

 inference: false
 ---
+⚠️ComfyUI's native int8 support prioritizes the old format, so this model does not work. Please continue using int8-fast.
 A Krea 2 model quantized to INT8 with ConvRot using a conservative quantization policy.
 <img src="ComparisonA.webp" width="1000">