Issues
- 0
Unable run workflow
#55 opened by BenderBlender - 0
Make confilct with really ComfyUI-IC-Light
#54 opened by sunhaha123 - 0
[DevTask] Flux IC-Light support
#53 opened by huchenlei - 3
KSampler Given groups=1, weight of size [320, 4, 3, 3], expected input[2, 8, 96, 96] to have 4 channels, but got 8 channels instead
#51 opened by Sinyuk7 - 3
- 2
Given groups=1, weight of size [320, 4, 3, 3], expected input[4, 8, 135, 240] to have 4 channels, but got 8 channels instead
#48 opened by TostormZ - 0
It's the same 'intermediate_dtype' function error that was there before, and it's happening once now.
#52 opened by flybirdxx - 1
TypeError: calculate_weight() got an unexpected keyword argument 'intermediate_dtype'.
#50 opened by StableQuestion - 11
Still getting !!! Exception during processing !!! Given groups=1, weight of size [320, 4, 3, 3], expected input[2, 8, 116, 128] to have 4 channels, but got 8 channels instead
#46 opened by talhadar90 - 3
Given groups=1, weight of size [320, 4, 3, 3], expected input[2, 8, 96, 96] to have 4 channels, but got 8 channels instead
#49 opened by JimWang151 - 4
KSampler error (like everyone else)
#40 opened by vitalbuttering - 1
- 4
Given groups=1, weight of size [320, 4, 3, 3], expected input[2, 12, 96, 64] to have 4 channels, but got 12 channels instead
#47 opened by zmczmc123654 - 8
ic-light was completely disabled when updating ComfyUI torch to 2.4.0, now it doesn’t work even in version 1.5 =(
#44 opened by ffdown - 1
Error occurred when executing KSampler: calculate_weight() got an unexpected keyword argument 'intermediate_dtype'
#45 opened by flybirdxx - 9
Layerdiffuse node installed. still get RuntimeError: Given groups=1, weight of size [320, 4, 3, 3], expected input[2, 8, 86, 128] to have 4 channels, but got 8 channels instead
#31 opened by nux1111 - 4
Error occurred when executing KSampler
#27 opened by TostormZ - 0
Add noise offset IC-Light model
#39 opened by fAIseh00d - 1
PytorchStreamReader failed reading zip archive: failed finding central directory
#38 opened by Amazon90 - 1
RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 8 but got size 128 for tensor number 1 in the list.
#35 opened by xm1085 - 0
how to deal with this issue
#36 opened by zmczmc123654 - 1
Cannot get image for Apple M1 chip
#34 opened by REDact324 - 0
Foreground changes after workflow
#33 opened by pandayummy - 1
"Header too Large"
#32 opened by revolvedai - 3
Error reported when adding UNETLoader
#13 opened by nftlaolong - 0
- 0
Why are the denoise level set to 0.9 when it's creating from an empty latent image?
#29 opened by ynie - 0
Weight Apply Issue (IC Light is currently unusable)
#28 opened by logtd - 1
'cc_projection.weight'
#23 opened by a2489805265 - 4
Error occurred when executing KSampler: Given groups=1, weight of size [320, 9, 3, 3], expected input[1, 8, 126, 94] to have 9 channels, but got 8 channels instead who knows how to solve this problem?
#17 opened by Jaylorandy - 0
AnimateDiff support
#24 opened by EvgenyKashin - 0
Unable to re-run workflow (WARNING SHAPE MISMATCH diffusion_model.input_blocks.0.0.weight)
#22 opened by dencs08 - 1
Error occurred when executing KSampler
#21 opened by epeiuss - 7
Error occurred when executing KSampler:
#18 opened by Onionman61 - 5
Error occurred when executing KSampler:
#19 opened by Chinnzzz - 0
Another Ksampler runtime issue...
#20 opened by MaestroFury - 0
- 2
Error occurred when executing UNETLoader
#14 opened by taylor6iu - 6
Error occurred when executing KSampler
#15 opened by rongtian0927 - 3
可能出了一些问题,关于fg.json主体颜色问题。
#12 opened by newxhy - 3
- 1
- 2
这个节点我在ComfyUI的Manager里怎么找不到?
#9 opened by angelussun - 6
Combining FG and BG results in correctly relit FG but the BG is not added back in.
#7 opened by paulerbear - 3
custom light map error
#6 opened by AugustRush - 4
输入512,4通道,采样器报错
#2 opened by 807502278 - 2
why?
#3 opened by mylonesky-bot - 0
- 1
Error occurred when executing KSampler: Given groups=1, weight of size [320, 4, 3, 3], expected input[2, 8, 96, 64] to have 4 channels, but got 8 channels instead
#1 opened by shumoLR