I just downloaded Comfy UI, here goes my weekend...
@pixaroma7 сағат бұрын
Have fun 😁
@defiantfaith3249 сағат бұрын
hello, why in itools prompt styler my negative text have to input value, thanks?
@pixaroma6 сағат бұрын
Check the discord post screenshot and explain what exactly you try to do, i think i saw you posting there
@Happipaws9 сағат бұрын
How do you have all your PC live metrics on display at the top of the screen?
@pixaroma9 сағат бұрын
There is a custom node you can install from manager, custom nodes manager, the node :crystools
@dioputra289314 сағат бұрын
hi can i know python_embeded python version u use please :)
@pixaroma12 сағат бұрын
Python version: 3.11.8
@dioputra289312 сағат бұрын
@ thank you
@PatrickLavender15 сағат бұрын
How did you make the song at the end, and how long did it take you to make that song? Was that AI generated also? Pretty catchy tune IMHO
@pixaroma15 сағат бұрын
you can use udio or suno,I used udio but now suno has v4 thats quite nice, so yes I created the song with ai, is quite easy just needs a few tries to get it right, like with ai images it needs a few generations and prompt adjusting to get what you want. If you want to see more examples you can check my other channels www.youtube.com/@altflux or www.youtube.com/@AI2Play
@Neveen.SalahAl-DinКүн бұрын
This is the first video I saw on the channel and I really want to thank you for the detailed tutorial and the amazing styles u provided 🙏
@GuysLightburnLaserGuidesКүн бұрын
Great video bro
@FacsHitDifferentКүн бұрын
hi amazing video in theory it should work but i get this error message everytime: dualcliploadergguf `newbyteorder` was removed from the ndarray class in numpy 2.0. use `arr.view(arr.dtype.newbyteorder(order))` instead. maybe someone know how to fix that really appreciate your help.
@pixaromaКүн бұрын
You can try this Downgrade NumPy to a version under 2 example go to ComfyUI_windows_portable folder, in the address bar type cmd and press enter then run this command, and restart comfyui .\python_embeded\python.exe -m pip install numpy==1.26.3
@FacsHitDifferentКүн бұрын
@@pixaroma thank you so much worked out. u do the best tutorials
@b3p_fl3xy88Күн бұрын
i cant get the right bottom thing that box which has save load refresh thing how to get that
@pixaromaКүн бұрын
That is in the old interface in the new one you have menu on top left where you have load workflow and refresh
@MissMisha-yx1gbКүн бұрын
trying to do in flux ui and always give error
@MissMisha-yx1gbКүн бұрын
any idea if we can use this in flux?
@pixaromaКүн бұрын
I havent used a1111 for months, i switched to comfyui, so i have no Idea what a1111 cand still do or not.
@danyoh1433Күн бұрын
awesome - thanks a lot!
@yapyh2872Күн бұрын
This version of LoRa loader is amazing, i don't have to go back to the website to check the trigger words and can manage multiple LoRa at the same time. Btw, i also want to suggest content for your consideration on how to create consistent character using character reference/sheet that can be done through controlnet that captures different angles and expression of a character. I saw it somewhere but forgotten. It would be really interesting to see your way of doing it.
@yapyh2872Күн бұрын
Cool. which ai voice are you using?
@pixaromaКүн бұрын
VoiceAir and they have the voices from elevenlabs. Voice is called burt us
@karamsahoo6852Күн бұрын
Class apart!
@paulcurry8383Күн бұрын
I'm suprised you were able to get flux to work immediately. I get AssertionError: You do not have CLIP state dict! which seems to imply I need to install other things for the model to work.
@pixaromaКүн бұрын
Make sure you are using the right t5 and clip models, they changed the interface after i done tutorial, so you need the right clip models for the right flux model
@icedzinniaКүн бұрын
When I did this tutorial, it didn't work (5:52). I realized after much fiddling around, using different loaded images, different masks, that it won't work if the mask is not continuous. I couldn't make it work when i had a mask on the far right edge and a seperate one on the far left edge. WEIRD.
@pixaromaКүн бұрын
I usually do one selection, there is on the inpaint crop an option called fill_mask_holes, maybe try to turn it off to see if helps, if not maybe is some bug
@AlloYa-h4pКүн бұрын
Wow SDXL beats the crap outta SD3. Also any censorship should be boycotted!
@pixaromaКүн бұрын
the new sd 3.5 medium is a little better, but still not as good as flux
@GreggoryLannisterКүн бұрын
Glad I found this, very concise and well thought out tutorials, just what I needed. Thanks
@AlloYa-h4pКүн бұрын
So simple, yet it looked so complicated, thanks a lot.
@pixaromaКүн бұрын
thank you 🙂
@AlloYa-h4p2 күн бұрын
Cool stuff, very helpful
@AlloYa-h4p2 күн бұрын
Thank you
@achidavitadze21812 күн бұрын
Hi,i was completelly beginner and after watching ur videos ive got something.. where can i download this workflow shown in the video?
@pixaromaКүн бұрын
on discord look for the pixaroma-workflows channel I have there for each episode all the workflows discord.com/invite/gggpkVgBf3
@TK-xn3bq2 күн бұрын
Just installed and followed your instructions, thank you! The interface that I have doesn't have the "Interrogate Clip". DPM++ 3M SDE Karra no longer is available. What do you suggest is the best option in place of that?
@pixaroma2 күн бұрын
I am not sure, maybe 2m instead of 3m sde karras or 2m karras
@TK-xn3bq2 күн бұрын
@@pixaroma Thank for the quick reply. This is all so new to me and took a couple of days for me to install it. It's working but very slow on my PC. I did change the file to make it run faster like you suggested in your other video but that didn't work so I had to undo that part. I basically want to turn some images into watercolor for a huge project. SD is definitely very interesting especially I haven't found any Action or filter from Photoshop to make this work.
@nsfwCharlie2 күн бұрын
The voice is so hard to listen to. Fucking AI generated shit. Give me a real fucking person any fucking day. I hate this crap.
@frankiberlin2 күн бұрын
Thank you so much! This must have been a lot of work. And it helped me so much to understand. I'm glad I found your channel. Cheers 😊
@pixaroma2 күн бұрын
thank you 🙂
@Amr_Cinema2 күн бұрын
alwayse the best
@GustavNielsen-b8x2 күн бұрын
Love this keep up the good work. Like the more detailed explanation!
@michielwes12 күн бұрын
Hi there, where can I find the Easy-Use nodes?
@pixaroma2 күн бұрын
go to manager, custom nodes manager, search for ComfyUI Easy Use and install it, by user yolain
@59Marcel2 күн бұрын
I'm in awe at the simplicity of the work flow and the fantastic results. Thank you for a very interesting and well explained tutorial.
@pixaroma2 күн бұрын
Thanks marcel ☺️
@emergentcomics492618 сағат бұрын
I couldn't agree more. Super easy to add to any workflow and amazing results.!
@darkenpain79552 күн бұрын
Please make FOOOCUS professional tutorials
@pixaroma2 күн бұрын
Sorry but with comfyui i have more control on how i can do things, focus should be easier i dont think it needs too many tutorials to figure it out
@darkenpain79552 күн бұрын
@pixaroma It means that Comfy UI is better? And it thinks that my 1660ti 6g graphics can handle it. It doesn't work at all in Fooocus. Thanks
@pixaroma2 күн бұрын
Is better for me for what i need, each interface is fit for someone depends on what you need. I need control and to be able to do what i want and do my own workflows so comfyui is good for me, but if you need something simple foooocus can be ok for you. Also your video card is not so good so maybe sd1.5 and some sdxl.can work so you can probably can not do more complex stuff with flux and control net
@darkenpain79552 күн бұрын
@@pixaroma Thank you very much for your help. My last question is, is the 3060ti 8g a suitable option?
@pixaroma2 күн бұрын
@@darkenpain7955 more ram the better, i will go with an rtx 16gb of vram if you plan to work with flux
@pawelthe16062 күн бұрын
Great video as always. Big thumbs up. Thanks
@KimikimiYKim2 күн бұрын
Love the video ❤ Will you later create a tutorial on how to merge models in comfyui custom advanced or pro
@pixaroma2 күн бұрын
I am still learning things about merging and training, but didnt really need it in my workflow when i found so many models online
@TK-xn3bq2 күн бұрын
Anyway to do this on a Mac?
@pixaroma2 күн бұрын
not sure, only if you find version that work on mac, you can try to see if other managed to install github.com/lllyasviel/stable-diffusion-webui-forge/issues/314
@mrkdman2 күн бұрын
Just happened upon this and...WOW...this node is a game changer. Thank you for this, as I'm still trying to figure out the basics of ComfyUI. This definitely could work as a companion alongside Photoshop.
@jhonysteak55122 күн бұрын
The dark theme prevents the launch on my side, bad it interests me, we can change it otherwise? Ah yes, I do not have DPM ++ 2M karras as a choice. suddenly the tutorial is more days or I took the wrong version? I think it’s still too early for me since I just installed it, if you have sources or a simpler tutorial to understand what it tells I’m interested. I installed this one since I failed to install the other, stable broadcast 111 or something like that.
@pixaroma2 күн бұрын
I am using comfyui lately, forge has changed a lot in the last months, like in a week after i finished the tutorial it had a lot of new other stuff, so is hard to keep up with all the changes.
@Oxes3 күн бұрын
I dont get why mine does not work, i paint the area where i want a change, but it just keeps giving me the same image over and over again, am using a pony model
@pixaroma2 күн бұрын
maybe it needs an inpainting model, did you tried with the juggernaut xl model to see if that is work and is not the model cause
@talismanna3 күн бұрын
Magic. Thanks!
@omarsharod3 күн бұрын
hey thanks for the vid, is it possible to generate 10 bit/pixel art images or use a different model in the workflow that does it?
@pixaroma2 күн бұрын
If you find a good model that does that maybe, I didnt try
@OProgramadorReal3 күн бұрын
Another great video! Thank you! Have you already talked about IPAdapter or plan to? I think it’d be an awesome addition to the playlist!
@RedRojo2103 күн бұрын
I wonder how well it would do on tshirt designs with text made with flux. Is there an option to do this in bulk as well. Like load 10 images or a folder to have it process one after the other. Also lets say you were doing 2 or 4 images at a time would it remove the background for all?
@pixaroma3 күн бұрын
I put a workflow on discord for batch upload, it removes the background to all
@SumoBundle3 күн бұрын
Something very useful for me. To bad my system is not as fast as yours. Thanks for the video
@tsentenari43533 күн бұрын
I can understand why you want to save the mask separately, to combine it with the whole image, to be able to freely add or remove parts of the original image, say by drawing in Photoshop. But if I am going to save a transparent png anyway - why exactly would I want to save the mask separately, too? Isn't the mask sort of contained implicitly in the tranparent png? For example, when I want to edit the mask in Photoshop: Don't I get the same result by ctrl - clicking the thumbnail of my loaded image, to create a selection, and then clicking the mask icon? Is the result different from loading the seperately generated mask, or is it the same?
@pixaroma3 күн бұрын
There are software that doesn't have simple options to load a mask, or video editors and so on, but I didn't think about ctrl click i think is similar
@tsentenari43533 күн бұрын
So simple and handy! tyty
@Uday_अK3 күн бұрын
This method saves so much time compared to masking with the pen tool! 🎉 These days, Photoshop can remove objects, but it doesn’t always get it right. That’s where ComfyUI shines-it’s easier to use and gives way better results. Thank you! 😊✨
@pixaroma3 күн бұрын
Thank you ☺️
@baheth3elmy163 күн бұрын
Thank you very much for the valuable education!
@Fayrus_Fuma3 күн бұрын
So 2 minutes is the average to remove the background? And that's on a 4090? I'm scared to try how long it would take on my system.
@pixaroma3 күн бұрын
No, it takes 2 minutes first time to download what it needs, only one time, after that is 2-3 seconds
@Fayrus_Fuma3 күн бұрын
@@pixaroma I see, thx.
@BorisJosz3 күн бұрын
Great video! Your series is gonna help bring me up to date with this program super fast and I heartily thank you for it <3
@SebAnt3 күн бұрын
As always, thank you for the awesome tutorials and bonus PS tips 🙏🏼
@pixaroma3 күн бұрын
Glad you like them ☺️
@glaiveproductions29133 күн бұрын
Hi there, How do the same thing with Forge, when indtalled via Pinokio ? I don't find the Web ui folder in there. Regards.