Відео

Great work, thank you

What i want to know is, Comfy already support the Facewap in Controlnet and Reactor for FLUX?

Hey I'm getting this error while using IPAdapter "Mat1 and Mat2 can not be multiplied" Can you help me with this issue?

Thanks!

In my case neither the search in the workflow section nor the insertion of workflows as shown in the video works unfortunately. I've tried this in all major browsers and also had a fresh install of comfy portable but nothing fixed that. Does anyone else have this problem?

thank you so much sir!

Thank you for this! Amazing content as always.

Thank you for the timely updates. The question now is how to use the Crystools to monitor the comp resources?

You are an absolute legend. Could you make a video on how to deploy ComfyUI on a serverless platform like Runpod and call its API from some frontend? Or are you maybe aware of some tutorials on how to do that? Thanks

Gracias por tus videos ! I like this new interface

Thanks!

Thanks for another great video! 👌👍 In general I like the new layout but it still could be improved. 1. I like to put all of my data (models, loras, output images, etc) on on another drive to save space on the SSD where the main installation is. I'd like the option to save everything like that to the directory of my choosing. There is some functionality - you can specify image out put directory in the startup batch file. There is a yaml file in the main directory where you can redirect all the models. . However, I just discovered preview images are automatically saved in the ComfyUI/temp directory. I also ran across a directory where it stored pasted images, but I for got where it was (which is kinda the point here). The workflow handler is great but it still saves the workflows in the ComfyUI/user/default directory, something I discovered only by poking around in the various folders. 2. The new queue system is not as functional as the old one. It shows the final image which is kinda useless. If a particular run is ongoing it just shows a rotating arrow. The old system had a list of ongoing processes which you could delete if you wanted. Maybe I missed it, but again, it just shows it is not intuitive like the old system.

Do you know how to restore previously saved workflows from the previous UI? I switched the UI to this beta version, and all my saved workflows disappeared. It's not a huge deal, but it would be nice not having to track down dozens of workflows. Thanks for the video, and thanks in advance for helping with my issue

How can we specify the load workflow folder from the workflow loader? I have a ton of workflows in a folder on another drive.

Still can't figure out how to save a workflow as favorites. (In workflows management area)

Wow! Great quality of life improvements! Especially the fuzzy search. Trying to find the node I am looking for was always a massive headache, especially when you have lots of custom nodes. Also the way we can organize workflows now is very welcome too. A quick question. In the official help page it says that I can collapse a node (shrink it so only the title is visible) by pressing ALT+C, but this doesn't work for me. How can I fix this? Is it possible to assign my own hotkeys? For example, I would like to also assign a hotkey that invokes the node search function as I find it a bit tiring to double click in a blank space every time I want to add a node. Thanks for the helpful videos! :)

Very informative as always 👍

Nice Guide!

I have hated using Comfy for a while because the UI felt disorganized to me and my ADD cant handle everything that is happening visually. Watching this video has given me hope that I can use it now

Thanks for always sharing such a great information.

awesome

Oh thank you !, that was really helpfull, i was doing the excact opposite by trying to ba as short as possible by just adding keywords. Now i understand why i was struggling to get what i wanted. I was sure that there was something wrong in my prompts and now i no what to do!

I enjoy your channel but here I think you perhaps aren't being quite analytical enough. The long LLM prompt for example. how much was it responsible for the image? If so, how many of those words were having significant effect? You can test this by removing classes of words systematically, so first the pronouns then adjectives, then verbs etc. When assessing you must be careful not to mix up different with better. The test should be whether the image has improved in quality, not whether it has changed. I know this is subjective but a rough assessment is not too hard. With Flux there is a great variation from seed to seed so it is easy to get a random very high quality image which might not reflect the average output. So testing across a range of seeds is important.

Great video but what is missing is why there is a need for two different prompt boxes for clip and T5.. what is the difference.

Thank you so much for your videos. They are so helpful.

Loved it! Thanks!

Great explanation! Question-Can I setup ComfyUI in M2/16 GB MacBook? Thanks

Very helpful, thank you! And congratulation for the new house!

Man you are simply brilliant and a hidden gem.

Thank you so much , you earned a sub with your explaination. TOp notch stuff

Thanks mate, i was having so many problems with this!

I followed your links and do not see those example videos of the new frontend features??? One suggestion, in your video, your mouse clicks & navigation are so fast, they're hard to follow (required rewatching a few times). Edit: Finally found them by expanding one of the links. Again, your navigation is too fast. Had you taken just a couple seconds to point out what you are clicking on and what you're expanding, it would be much easier to follow your navigation.

Got this error every time Error occurred when executing PulidInsightFaceLoader: File "H:\ComfyUI_windows_portable\ComfyUI\execution.py", line 317, in execute output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "H:\ComfyUI_windows_portable\ComfyUI\execution.py", line 192, in get_output_data return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "H:\ComfyUI_windows_portable\ComfyUI\execution.py", line 169, in _map_node_over_list process_inputs(input_dict, i) File "H:\ComfyUI_windows_portable\ComfyUI\execution.py", line 158, in process_inputs results.append(getattr(obj, func)(**inputs)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "H:\ComfyUI_windows_portable\ComfyUI\custom_nodes\PuLID_ComfyUI\pulid.py", line 240, in load_insightface model = FaceAnalysis(name="antelopev2", root=INSIGHTFACE_DIR, providers=[provider + 'ExecutionProvider',]) # alternative to buffalo_l ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "H:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\insightface\app\face_analysis.py", line 43, in __init__ assert 'detection' in self.models ^^^^^^^^^^^^^^^^^^^^^^^^^^

Is there allready an XY script possibility for FLUX to test the LORA that I trained?

Thank you very much! You are best!!!

4:27 this worked for me earlier today! then i installed some lora training application from within comfy manager and i could no longer get back into comfy! so now im trying to reinstall from scratch and this time when I typed in that python code at 4:27 it told me: Installing collected packages: opencv-python-headless, onnx, lazy-loader, joblib, imageio, eval-type-backport, cython, annotated-types, scikit-learn, scikit-image, pydantic, albucore, albumentations, insightface ERROR: Could not install packages due to an OSError: [WinError 5] Access is denied: 'D:\\ComfyUI_windows_portable\\python_embeded\\Lib\\site-packages\\cv2\\cv2.pyd' Consider using the `--user` option or check the permissions. updatE: i went to the root folder on the pc and clicked properties and unchecked read only minus box and i think that fixed it. no clue why it was set to that

ok i think the WHL part of your instructions mightve fixed it for me thank you!

none of this is working for me. i tried to install through manager and got an error. then i went and installed the repo from github. still didnt work. then i opened my comfy window and typed in CMD and it opened a command window and I typed in LS and it told me D:\ComfyUI_windows_portable>ls 'ls' is not recognized as an internal or external command, operable program or batch file. but if i watch someone elses tutorial and type in python --version it says i have python 3.11.9

Great info. Thanks!

You have a great channel! I’m looking forward to more videos, especially around ComfyUI and Flux!

can you please explain if we have 2 loras, and they are different persons, how can we implement, it without effecting each other. lets think like a couple picture?

appreciate if you could adjust the bass sound down a bit, sound like you talking in tunnel....

I’ve been looking everywhere. Can I run all of this from my D drive so I don’t fill up my C drive?

this is error `decoderF` is not supported because: max(query.shape[-1] != value.shape[-1]) > 128 xFormers wasn't build with CUDA support attn_bias type is <class 'NoneType'> operator wasn't built - see `python -m xformers.info` for more info `flshattF@0.0.0` is not supported because: max(query.shape[-1] != value.shape[-1]) > 256 xFormers wasn't build with CUDA support operator wasn't built - see `python -m xformers.info` for more info `cutlassF` is not supported because: xFormers wasn't build with CUDA support operator wasn't built - see `python -m xformers.info` for more info `smallkF` is not supported because: max(query.shape[-1] != value.shape[-1]) > 32 xFormers wasn't build with CUDA support dtype=torch.float16 (supported: {torch.float32}) operator wasn't built - see `python -m xformers.info` for more info unsupported embed per head: 512

I get an error message when running the workflow: [F dml_util.cc:118] Invalid or unsupported data type Float_e4m3fn. Googling this lead me to running an update of PyTorch to v 2.4.2 but then the ComfyUI doesn’t even start at all anymore due to all kinds of missing DLLs etc… Maybe I am the only one with these issues?

Hi there!! I am following your guide on macOS. I am stuck at the Python Embedded. I don't seem to have that. Should I just skip it? Thanks in advance 😇

The Flux Community is ON fire. I am still not able to get great consistent images with Flux, its limiting in the body style controls of generating people. However, I hope that changes now with ipadapter and control net added to Flux.

You skipped and didn't mention Disco Diffusion.

@CodeCraftersCorner An odd question: with img2img, can AI forced to basically reproduce the original image? if so, which setting would achieve this?

Thanks for your clear and patient explanations! 🙏