BlenderNeko/ComfyUI_ADV_CLIP_emb
ComfyUI node that let you pick the way in which prompt weights are interpreted
PythonGPL-3.0
Issues
- 0
- 0
CLIP文本编码器(BNK-SDXL) The expanded size of the tensor (19096) must match the existing size (248) at non-singleton dimension 1. Target sizes: [-1, 19096, -1]. Tensor sizes: [1, 248, 768]
#33 opened by ZHUMI-CC - 0
list indices must be integers or slices, not str
#32 opened by Neytiri7 - 8
Is this working?
#31 opened by andupotorac - 3
- 0
Add support For Flux
#30 opened by Imadbein46 - 0
Pixart sigma support
#28 opened by ReyJ94 - 0
Are any workflow and json available?
#27 opened by CoUIfy - 0
I don't understand how I'm supposed to branch cutoff regions to BNK_CLIPTextEncodeSDXLAdvanced
#24 opened by Maveyyl - 6
SDXL model throws out clip_layer errors after ComfyUI commit c2cb8e8 (Feb 25, 2024)
#20 opened by SakanakoChan - 0
Fix bug when using sd2.0
#23 opened by zhangsdly - 13
- 1
Results are identical for all parsers
#18 opened by Graylington - 0
Stable Cascade Support
#19 opened by rethink-studios - 0
Weights below 1 are broken
#17 opened by ponchojohn1234 - 0
Include simple example workflow to show usage
#16 opened by ghostsquad - 0
- 4
- 3
- 1
How to leverage wildcards?
#11 opened by alenknight - 4
Comfy++ and A1111 parsers halt the workflow
#8 opened by lanzjose - 4
- 4
All options generate the same image
#9 opened by davemedvitz - 2
- 0
[Appreciation Awareness] Thank you
#4 opened by WASasquatch - 2
Tensor Size mismatch error
#5 opened by TinyTerra - 4