Відео

@@seriousShane thank you so much ✨

@@mikrobixmikrobix here is the trick , in the first node "Load diffusion model" for weight type choose e4m3fn and that horizontal lines will gone. It took me weeks to figure out

@@octopuss3893 Thank you my friend 🙏🏻🔥

Got a little problem: can't find the "AIO Aux Preprocessor" node 🤔

@@simuf563 you're welcome. Did you download my workflow?

@@Skyrim973 yes I tested it and it has slightly better quality than dev model. It's up to you. But about paying I recommend using the Replicate website. It is the most efficient and cheapest platform to use flux pro ultra. I guess that was 0.04 per image. But not sure check it yourself.

You're the MVP! 🙌🔥 Appreciate the support!

Select manually that flux1-redux-dev.safetensonrs from Load Style Model node

what is the error message?

@@Jockerai StyleModelLoader: invalid style model. I replied with the solution yesterday, but effing utube deleted it. I'll try posting it again.

​@@Jockerai If you get the StyleModelLoader: invalid style model error. The solution is to 1. Make sure you have the latest version of Comfyui installed (I don't know how to do this). 2. Go to the ComfyUI_windows_portable "update" folder and run the batch file: update_comfyui_and_python_dependencies.bat. In my case it downloaded 2.4 GB and updated a bunch of ComfyUI and Python dependencies. I then had the new ComfyUI version and the workflow ran without an error. Thanks @Jokerai for all the great content, and thanks utube for making me post twice.

@@MrWill-ng8dg you're very welcome my friend ✨. Thank you for sharing this solution.

@@bartosak you're welcome my friend ✨

You're welcome my friend ✨ Yes it does. You can see in the video in the lora loader node , the 8 step turbo lora is active. All workflows for this tutorial has this capability

Thank you so much my friend that was a really uplifting compliment I appreciate that. Yes in the tests I have done you don't have to use trigger words.but sometimes it can affect your output

You're welcome my friend, Thank you for your attention and kind message!✨🤗

Even with 6 VRam you can use it. If you have less watch my video which has titled Flux with 2vram and use GGUF nodes in this workflow instead of main flux model node( the first and second nodes)

Glad I could help my friend ✨ it is one of the best capabilities of Redux and I was really surprised by that and love it

You are so welcome!

I fixed that. Click on the links for downloading workflows in the description

@@Paulo-ut1li you're welcome my friend ✨. lora creation : ua-cam.com/video/HNtTC0oo9I0/v-deo.html . Flux new Inpainting is the next video which is very new for flux and it has unbelievable quality stay tuned 😉

You're welcome 😊

@@TheBlackbeansoup you're very welcome my friend ✨. What if I say it has a V2 which is more detailed and has more improvements? :) Check it out: ua-cam.com/video/T4SNWa-izxM/v-deo.htmlsi=uLGC6Ftccw84zJ6P

simply V1.0 is an update from V0.3. V1.0 doesn't need VAE and it is more optimized and has better train

Thank you so much for your kind and encouraging words! I'm glad you found the tutorials helpful. Wishing you a successful, joyful, and wonderful 2025 as well! 🌟

Please watch version 2 of this tutorial it can be more usfull for you and no issues : ua-cam.com/video/T4SNWa-izxM/v-deo.html However the this version is also great.

@@scratch45 Hi bro I know that and you can use GGUF versions of flux. Simply watch my video in this link :ua-cam.com/video/0plLyIp1xbI/v-deo.html . Then open this upscale workflow and replace the first and second nodes (Load diffusion model and Dual Clip Loader) with GGUF nodes (Unet Loader GGUF and Dual clip loader GGUF). Don't forget if your system is weak you should use Q4 or Q5 models of GGUF models which I explained in that video I shared the link. Hope this can help

@Jockerai Thanks bro 🎩 I appreciate that for what you are doing and keep going 🩶

@@scratch45 Thank you so much my friend

@@researchandbuild1751 there is a new Inpainting method which I will make video for that. Indeed it will be V2. Which I said in my last short

@@АндрейАлександрович-ч7ж check second nodes (Dual clip loader). Third option should be flux not Sdxl

@@JY-gf9zi it can be used but most of the time I don't use that

@@tat98napopova92 It should run this workflow with no problem. You should send me your cmd log. Dm me on telegram or Patreon

به سختیش فکر نکن مراحل رو یکی یکی انجام بده سختیش یه روز اوله. بهتر از اینه که بری پول به سایت ها بدی هر ماه و حتی امکانتشون هم کمتر باشه.

@@philippeheritier9364 thank you mate you're very welcome my friend ✨🫂

@@ceegee9064 you're very welcome my friend 🙏🏻

@@chasingdaydreams2788 " Thank you so much bro for your support and positive comment! I'm truly grateful to have viewers like you. You're absolutely right; instead of spreading negativity, we should focus on encouraging and uplifting each other. Your kind words really made my day!🤘🏻🫀

@@pjr_morstance_official Yes you can request it in this link : www.patreon.com/Jockerai/shop/1-hour-chat-assistance-with-jockerai-657031?Link&

@@sammehrani-q6h you are welcome bro🌱❤️

@@KaptonTimo you're welcome bro ✨

just follow the steps and you can make it easily. If you can't just dm me on Petreon or Telegram I will send you the workflow

@@Jockerai i do follow the steps ,but even i put models file in controlnet folder i still can't chose model on "load contrlnet note"

i find refresh buttom on new ui and solve this problem

if i made pose already(fix hand and off screen body) what note i shoud use?

and how can i load checkpoint on this workflow? thanks for anser