Видео

Could you please make another workflow for flux?

This has now been of course moved to the native Settings dialog > search for "badge" and press Enter > "Node ID badge mode" setting. It will surely be moved and renamed again by the time you read this comment.

Hi there, I am a bit confused on where the Image Receiver is getting its image from. I downloaded the workflow from the github page, however even when running, the "Image Receiver" is not loading any image. From my understanding, it should be coupled with an "Image Sender" node but it's not present in the workflow. The only thing I can think of is that it's somehow receiving the image from the "LoadImage" node, but mine doesn't behave like that and I'm not aware of any settings for enabling that behavior.

How do I connect two "sdxl prompt stylers" to use them at the same time?

get a mic

why no voice?

If I have a set of segments (Seg1, Seg2, Seg3) and I want a different prompt to apply to each of them, do I understand correctly that the Wildcard will be {prompt_for_seg1 | prompt_for_seg2 | prompt_for_seg3}, i.e. the wildcard pieces will be processed sequentially like the segments? And if not and the Wildcard pieces will be taken randomly, is there any way to assign each segment its own specific prompt?

Hello @drltdata, I miss your great ideos... Just a question please, is there a new or better way to let the detailer for AnimateDiff accept more than 30 images? I am using this workflow but I use Load Video (path) .. the video is 120 images.

it's not working for me for some reason, it downloaded the model and now it's just not displaying anything. Am i too impatient? how long should this take?

thank you man you saved me

for some reason, the queue load stops on the second region and loads infinitly

at 2:39 how could you show preview image on KSampler node? Do you have install any packages for that ?

I'm unable to find the MediaPipe-FaceMeshPreprocessor node anywhere, I did install the addon packs.

everything goes well until it reach toladapterpipe then it stop and "prompt executed" I really don't know what to do

certificate error

thanks.

Seems like it is blocked, I get this error when I tried to install the nodes: 'With the current security level configuration, only custom nodes from the "default channel" can be installed.' I am on mac and comfy is the most current one, I do not even know what 'security level' here means. Is it the only way to make and use colored mask?

Is it possible to get XY coordinates of each seg?

I can't see "Badge" :/ Everyhing has been updated.

DOES NOT WORK FOR NESTED WILDCARDS. aside from zero proof, and a claim on their github, it does not. I've tested it in the latest iteration of comfyui. DOES NOT WORK!

Does the Impact-Pack support nested wildcards?

thanks for the video! is there a way to do this without inpainting? just to control attention mask before ksampler? (noob here :P)

is there a way to use combined, but make the masks as close to each other as possible to use detailer? In my case, I have masks of head, left hand, right hand, and two legs, I want to inpaint those areas, but those masks are very far way from each other, making the size of the image very large. I want to bring all those mask to close to each other, combined to one image, upscale/inpaint, then use special mapping to paste those refined parts to the original image

so you are responsible for this amazing workflow thanks, could you tell us where is it located on the drive, to delete or share with people Thank you for your amazing work, subbed

I've learned that just waving my mouse cursor around the comment entry line does nothing to teach you about how I loath silent videos and that if you have an aversion to microphones it isn't that difficult for you to install an ai voice generator to explain things like a normal person. If all you do is wave your mouse cursor around without explanation in a video where most of the words aren't terribly legible most of the time, you may as well just make a short 5 second video that's nothing more than a web address to a large readable image of your workflow.

Thanks! Have to say that without comfy manager the whole user experience in ComfyUI wouldn't be so nice as it is now.

why do my final characters look nothing like my inputs?

hi. i just want to know if you are going to add more hairstyles. can we add more ourselves?

Why didn't you use the actual lora, while checking it with XY Input: Lora Block Weight? I can't see it loaded in the Efficient Loader.

please talk friend, your videos could be amazing

Awessome! Thanksss

basically crop_factor & bbox_fill both provide more context to the detailer, correct?

What is the model name ?

Where I can find this workflow?

My name is Joseph Jerry. I just finished watching your video, and it was very insightful. However, I'm having trouble downloading ComfyUI from the direct link on GitHub. Could you please help me understand what the problem might be?

很少看到有人介紹.. 謝謝你.

I have tried everything but no image preview after queue prompt on my rtx 2070. (i have no issue with my rtx 3090)

Nothing compares to your extension! Thank you immensely for all you do!!!

thanks

Hi the two github links in the description aren't working are you able to reupload them?

Very clean workflow, I find it strange some connections seem to not have what I'd define as a "conclusion." like the Lora clip output. But im still wondering how are you able to watch the image generate like that?

Thank you for sharing. How to put photos from the outside?

Doesn't seem to work with Inspire Pack version_code = [0, 82], loaded your workflow, tried increasing the strength a lot, change the seed, made the mask a lot bigger, nothing changed, can you take a look. Thanks for all the custom nodes, they are very useful. Edit: did more testing, found out if noise mode is set to CPU for all nodes, the Seed Explorer By Mask doesn't work.

he tell error while installing wd v1 moat tagger any solutions

Could you give a link for this workflow? I still don't understand how it work but I think this can solve my problem.

it's not a tutorial, it's you moving your screen around.

Incredibly helpful for getting things up and running from an Auto1111 mindset. Thank you!

tk anh đã chia sẽ nhưng mình bị lỗi ksampler khắc phục thế nào ạ tk anh nhiều

thanks for share👍

i downloaded this workflow, while "2sampler for mask" is working as shown in the demo but "2 advance sampler for mask "is not working ..after Q prompting all its giving is noise , where im doing wrong ?? any clue??