TencentQQGYLab ComfyUI-ELLA issues

TencentQQGYLab / ComfyUI-ELLA

ELLA nodes for ComfyUI

334 stars 14 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

AWPainting-v1-2 not find . in examples

#64 bobby20180331 opened 1 week ago
0
希望开源ELLA SDXL版

#63 leohu1 opened 1 month ago
1
Unable to load Flan-T5 Text Encoder

#62 Kasr40k closed 1 month ago
5
Token indices sequence length is longer than the specified maximum sequence length for this model

#61 darkflare opened 1 month ago
0
fix: property `device` has no setter

#60 JettHu closed 1 month ago
0
On a mac: Error occurred when executing T5TextEncoderLoader

#59 andupotorac closed 1 month ago
4
Error occurred when executing T5TextEncode #ELLA:

#58 jinwen36 opened 1 month ago
10
ComfyUI web interface stops open (endless load)

#57 DaveScream closed 1 month ago
1
Can't set attribute "device"

#56 atxoxx opened 2 months ago
15
Extremely slow generation

#55 farvend opened 2 months ago
2
T5加载器问题

#54 DuanMuChaoFan opened 2 months ago
2
Unable to load vocabulary from file. Please check that the provided vocabulary is accessible and not corrupted.

#53 chrislailiru opened 3 months ago
2
文件路径错误

#52 xm1085 opened 3 months ago
1
Error occurred when executing T5TextEncoderLoader #ELLA: It looks like the config file at 'D:\ComfyUI\ComfyUI_windows_portable\ComfyUI\models\ella_encoder\models--google--flan-t5-xl--text_encoder\config.json' is not a valid JSON file.

#51 AngryOldWang closed 3 months ago
0
Worse image quality when combine with ipadapter

#50 Scorbinwen closed 4 months ago
2
用了ComfyUI-ELLA反而更加不听说了，而且质量更差

#49 zoenders opened 4 months ago
1
rror occurred when executing T5TextEncoderLoader #ELLA: Error(s) in loading state_dict for T5EncoderModel: size mismatch for shared.weight: copying a param with shape torch.Size([32128, 2048]) from checkpoint, the shape in current model is torch.Size([32128, 512]). size mismatch for encoder.block.0.layer.0.SelfAttention.q.weight: copying a param with shape torch.Size([2048, 2048]) from checkpoint, the shape in current model is torch.Size([512, 512]).

#48 flyricci closed 4 months ago
1
Question: Using Embeddings in T5 Text Encode Node

#47 Architech-Eddie closed 4 months ago
1
Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu!

#46 teanhow opened 4 months ago
1
AYS + ELLA + AnimateDiff anyone?

#45 Duemellon opened 4 months ago
1
请问大佬什么时候更新XL模型啊？

#44 qingdengke88 opened 4 months ago
1
M1 max, Error occurred when executing EllaApply: "LayerNormKernelImpl" not implemented for 'Half'

#43 yiwangsimple closed 4 months ago
0
Error occurred when executing T5TextEncode #ELLA: (RX580 i39100f Windows11 32gb Ram)

#42 KillyTheNetTerminal opened 5 months ago
6
Update exsamples

#41 JettHu closed 5 months ago
0
你好，RuntimeError: mat1 and mat2 shapes cannot be multiplied (89x1024 and 2048x768)

#40 kongyingzhe opened 5 months ago
1
AYS + Ella = anyone else getting superdark image generations?

#39 Duemellon closed 4 months ago
8
load ella model

#38 one-pip opened 5 months ago
1
load ella model

#37 one-pip closed 5 months ago
0
Apply ELLA node error，TypeError: cannot create weak reference to 'NoneType' object

#36 whmc76 opened 5 months ago
2
feat: add `ELLA Text Encode` node

#35 JettHu closed 5 months ago
0
T5TextEncoderLoader显示报错

#34 ranfengqaq opened 5 months ago
19
A1111 type prompt features?

#33 Duemellon opened 5 months ago
2
Error occurred when executing EllaApply: "LayerNormKernelImpl" not implemented for 'Half'

#32 andulegal closed 5 months ago
2
help！

#31 jokero3answer closed 5 months ago
1
AYS + ELLA

#30 seanphan opened 5 months ago
4
NSFW even with negative

#29 MaraScott opened 5 months ago
3
'ModelPatcher' object has no attribute 'get_model_object'

#27 uxyz-yang opened 5 months ago
9
feat: upgraded ELLA condition apply method

#26 JettHu closed 5 months ago
0
Could you please tell me if this plugin supports travel prompts? It's really amazing, and I'd like to use it for animatediff's video generation, with different prompts applied to different frames. Thank you!

#25 hahy36 opened 5 months ago
4
Using the new "Concat" node to combine totally empty "T5 Text Encode" with normal "Clip Text Encode" gives WAY better results than actually using "T5 Text Encode", Lora or no Lora

#24 Akira13641 opened 5 months ago
3
Please add drivable external input to prompt nodes.

#23 DavideAlidosi opened 5 months ago
1
feat: support multi batch, lora tirgger words, prompt weight with clip

#22 JettHu closed 5 months ago
0
批次生图时报错或者没有正常工作？

#21 Jacob-HYG closed 5 months ago
1
SDXL supported?

#20 za-wa-n-go closed 5 months ago
1
RuntimeError: "addmm_impl_cpu_" not implemented for 'Half' when executing T5TextEncode #ELLA

#19 Akira13641 closed 5 months ago
2
Load T5 Text Encoder dropdown is empty

#18 eliganim closed 5 months ago
2
Anyone else having a "realism" issue with generations?

#17 Duemellon opened 5 months ago
3
Concat conditionings

#16 m0rtus59 closed 5 months ago
6
fp8 is not supported

#15 emptybamboo opened 5 months ago
6
Output image quality is unstable

#14 PetitPrinceb612 closed 5 months ago
4