-
-
Notifications
You must be signed in to change notification settings - Fork 287
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
🐞 Common issues. Please read! #108
Comments
@dustintheweb on what commit number are you on with ComfyUI? I see you are on linux (or Mac?), use git pull to update, not the manager. Then stop comfy and restart. Your installation is not up to date, even if the Manager says it is. |
@dustintheweb can you please post line 41 of your local file |
have you stopped comfy and restarted? |
Yep, all the things 🤷 |
I don't know if it's a Mac thing, I have no way to check unless you give me SSH access to your PC. The only other thing I can suggest is to start over with a clean installation |
[Load CLIP Vision ] is wrong |
ah right that's another possibility |
downloaded these from huggingface & updated the CLIP vision file as the only change, now have his error at Load Checkpoint:
![]() |
I would try to recreate the environment from scratch |
yeah it seems like something broke in general from an update this morning. I will nuke all and rebuild from scratch today / tomorrow and follow up. ty |
can you try to force fp32? |
mps should be fixed now |
Appreciate the support, @cubiq. Despite updating the code, I encountered a snag related to the dtype specification:
To tackle this, I explicitly set the dtype in the init() to torch.bfloat16, seems to be working since, dunno if it's the correct fix for all tho. |
can you try the latest commit? if it doesn't work please open a new issue |
this fixed the |
Hello .. Anyone have an idea how to resolve this error please? Error(s) in loading state_dict for Resampler: File "/home/admin/ComfyUI/execution.py", line 153, in recursive_execute |
`Error occurred when executing IPAdapterApply:
This afternoon's comfyui update (ComfyUI: 1805340177 Manager: V1.11.1) plus following the advice above to install into separate 2 folders in the /ComfyUI/models/clipvision of the safetensors below (both files have the same name hence why I created SD & SDXL folders to hold each) https://huggingface.co/h94/IP-Adapter/resolve/main/models/image_encoder/model.safetensors I can select either model.safetensors in the Load CLIP VISION node ... Thanks to you for all this amazing work. It opens a whole new world to me |
just download the missing model in the link https://huggingface.co/webui/ControlNet-modules-safetensors/tree/main |
clip missing: ['clip_l.logit_scale', 'clip_l.transformer.text_projection.weight'] e is external drive, please help!! |
Error Info: CrossAttentionPatch.set_new_condition() got an unexpected keyword argument 'cond_alt' File "I:\AIallinone\ComfyUI-aki\ComfyUI-aki-v1.3\execution.py", line 151, in recursive_execute |
IPAdapter Model Not FoundI have tried all the solutions suggested in #123 and #313, but I still cannot get it to work. Attempts made:
Are there any other solutions? I would greatly appreciate any help! |
U can use " ipadapter model load " to instand of "unified load", and Can you find model files in " ipadapter model load "? |
Thank you for your suggestion! |
it seems u should import " load clip vision" node, and connect to you ipadatper |
Thank you for your response. I actually managed to solve the issue myself later on and forgot to update the thread. I ended up using the Load CLIP Vision and IPAdapter Model Loader methods as you suggested, and everything worked perfectly! Thanks again for your help! |
Error occurred when executing IPAdapterAdvanced: local variable 'face_image' referenced before assignment File "/container/workflows/default/ComfyUI/execution.py", line 151, in recursive_execute |
|
Error occurred when executing AnyLineArtPreprocessor_aux: An error happened while trying to locate the file on the Hub and we cannot find the requested files in the local cache. Please check your connection and try again or make sure your Internet connection is on. File "E:\Comfy UI\ComfyUI-aki-v1.3\execution.py", line 152, in recursive_execute 求解,谢谢 |
Error occurred when executing IPAdapterFaceID: IPAdapterAdvanced.apply_ipadapter() missing 1 required positional argument: 'model' File "G:\ComfyUI-aki-v1.3\execution.py", line 152, in recursive_execute what should i do,thk |
I have fixed the error: Error occurred when executing IPAdapterUnifiedLoader: ClipVision model not found rename the files in the clip_vision folder as follows |
|
Before posting a new issue, please check the currently opened and closed issues! Very likely the solution is already there!
The most common causes for issues are:
Always update ComfyUI and the IPAdapter extension to the latest version. If you are on windows you may need to re-download a new portable version or use the update scripts!
After the update always stop ComfyUI and restart it. Then refresh the browser a couple of times to clear the cache. If it doesn't work try to recreate the updated nodes.
All the following errors are caused by an outdated installation:
The three technologies are very close together and share some common code. Be sure to upgrade all of them before reporting an issue.
You may have already installed the deprecated
IPAdapter_ComfyUI
extension. That will conflict with this extension and needs to be removed.The
IPAdapter Apply
node is now replaced byIPAdapter Advanced
. It's a drop in replacement, remove the old one and reconnect the pipelines to the new one.Any tensor size mismatch error is caused by the wrong combination of IPAdapter model, image encoder and/or base checkpoint.
All
-vit-h
models require the SD1.5 image encoder. At the moment only one SDXL model and thevit-G
SD1.5 model need the bigger image encoder.If you use any FaceID model you need either use the simple
IPAdapter
node or the dedicatedIPAdapter FaceID
node.The embeddings are saved into
output
directory and need to be moved into theinput
directory to be loaded.When you download the encoders from huggingface they both have the same name (
model.safetensors
). Please be sure to rename them correctly (for sd1.5 and sdxl) and use the right one depending on the IPAdapter that you use.All IPAdapter models use the "SD1.5" image encoder (no matter the target checkpoint) except for one SDXL model and models ending with
vit-G
.If you get errors like:
Expected query, key, and value to have the same dtype, but got query.dtype: struct c10::Half key.dtype: float and value.dtype: float instead.
Run ComfyUI with
--force-fp16
The text was updated successfully, but these errors were encountered: