Відео

Ask AI Search this guy doesn't know shit about this, he stole the script and video from AI Search. btw you can delete this comment also, but i will do a video about it.

I used an RTX 3060 with 12GB of VRAM, which achieved an estimated generation time of approximately 20 minutes per run. I believe you can use 8GB of VRAM, however the waiting time could be upto an hour or more which may not be ideal for your workflow.

@@AiMotionStudio Thanks for your reply. One more question: how much space does the installation take up on the hard disk?

@@gunther6638 Around 20GB was used in space.

You are welcome glad that you liked it!

I believe that it is not possible due to the fact that you need an NVIDIA CUDA GPU enabled to run this tool!

Exactly!

make sure you use CMD into the path specified on the tutorial, if your path is not correct you will get this error.

Make sure 1- ComfyUI is uptodate. 2 - Goto the "...ComfyUI\custom_nodes\ComfyUI-PuLID-Flux \" Directory. 3- On the address bar type "CMD" to go into Command prompt. (Black screen with current directory) 4 - On the prompt type "pip install -r requirements.txt" Make sure the python is installed. 5- After few minutes depending on the speed of the internet it will download everything and install the dependencies and create directories in COmfyUI. 6 - Type "pip list" and check the list if the items are listed in requirements.list "facexlib, insightface, onnxruntime, onnxruntime-gpu, ftfy, timm"

Yes i am working on a workflow for the Flux Tools. we will release very soon thanks!

Try to update the comfyUI to the new 2.0 version this should solve the errors

@@AiMotionStudio I redownloaded ComfyUI, and re-did the whole process. It is Comfy 2.0, still same error :( Maybe it's the lack of vram? I have a 2080 Super. I tried using GGUF models as well.

We were able to fix this error using this: github.com/comfyanonymous/ComfyUI/issues/5601 just go to your E:\pinokio\api\comfyui.git\app\comfy\ldm\flux folder and edit the model file with notepad+ and search for "out = self.forward_orig" line 184 and remove ", transformer_options" at the end of that line and save changes this will fix this errors

Yes how ever it will take a very long time I will suggest that you use flux Pulid instead. Check out our Flux Pulid video tutorials

Yes that’s true thanks for sharing!

Use a lower denoise value

before you accept the changes you need to right click on the edited image and save the image before accepting the changes.

the first time you generate it will take long because the models are being generated, but after that the next generation take around a mins depending on the MAC M1,2,3,4 you are using.

yes we then trained that voice in eleven labs

you need to bypass the florence2 node, U-Nai Get, and add CLIP Text Encode (Positive Prompt) and link it to FluxGuidance conditioning.

actually you can just add in the emotion in the prompt and it will be applied.

Just the installation part wink wink

not stolen but reused got inspiration from it!

No it didn’t 💀

Stolen and changed voice with AI for real

@ it isn’t stolen because their video literally released first 💀

@@respectvetman what are you talking about?? AI SEARCH posted one day earlier

Make sure 1- ComfyUI is uptodate. 2 - Goto the "...ComfyUI\custom_nodes\ComfyUI-PuLID-Flux \" Directory. 3- On the address bar type "CMD" to go into Command prompt. (Black screen with current directory) 4 - On the prompt type "pip install -r requirements.txt" Make sure the python is installed. 5- After few minutes depending on the speed of the internet it will download everything and install the dependencies and create directories in COmfyUI. 6 - Type "pip list" and check the list if the items are listed in requirements.list "facexlib, insightface, onnxruntime, onnxruntime-gpu, ftfy, timm"

make sure your path is correct dont use mine, Mine is drive E: maybe yours will be C: please check to make sure!

@@AiMotionStudio there isnt python.exe in scripts should I have it ?

​@@AiMotionStudio after having some problems I reinstalled python and comfy I have to install pulid rn but when I try to install for example C:\Users\user\Desktop\ComfyUI_windows_portable\python_embeded\Scripts\python.exe -m pip install C:\Users\user\Desktop\ComfyUI_windows_portable\ComfyUI\insightface-0.7.3-cp312-cp312-win_amd64.whl onnxruntime this insight face I get this C:\Users\user\Desktop\ComfyUI_windows_portable\python_embeded\Scripts\python.exe' is not recognized as an internal or external command,operable program or batch file. same for requirements.txt as well

AI made this bro Chill!

i already did it, here is the video tutorial: ua-cam.com/video/KtYooOgW7t0/v-deo.html

Glad it helped, Welcome to the channel!

@@AiMotionStudio I get this error AssertionError: Torch not compiled with CUDA enabled

yes we used the installation as a inspiration for this video!

Yes it is a clone video from AI Search /:

Yeah copy paste lol from AI Search!

you can actually use 12GB VRAM as well !

@@AiMotionStudio Does it work well on 12gb still? I heard it needs 15gb.

@@TPCDAZ Yes it does!

@@AiMotionStudio Great thanks!

i will look into that thanks for the Tip!

most people find it working after they installed the latest update version of comfyui 2.0. i think the problem is with the old comfyui had not fixed the torch pip errors that usually comes with using comfyui. Try to check on the video again, follow all the steps and instal the latest Comfyui. hopefully it will fix the issues. Let me know if you have any other questions.

We were able to fix this error using this: github.com/comfyanonymous/ComfyUI/issues/5601 just go to your E:\pinokio\api\comfyui.git\app\comfy\ldm\flux folder and edit the model file with notepad+ and search for "out = self.forward_orig" line 184 and remove ", transformer_options" at the end of that line and save changes this will fix this errors

yes please watch our tutorial about that topic: ua-cam.com/video/xUduNl7-pE0/v-deo.html

you can set the Browser to darkmode as well in the browser settings to enforce the darkmode thoughout the browser.

here is the workflow: civitai.com/models/309713/animatediff-sd15t2vbeta-best-workflow-for-realistic-vision

Yes we used this method: ua-cam.com/video/KtYooOgW7t0/v-deo.htmlsi=Fpyhfx1u_KVdj4z9

the Loras should be authomatically be saved as .safetensors files check to see that you setup the fluxgym installation correctly.

maybe in the future update, you can use free LLM as well ollama 3.2

most people find it working after they installed the latest version of comfyui 2.0. i think the problem is with the old comfyui had not fixed the torch pip errors that usually comes with using comfyui. Try to check on the video again, follow all the steps and instal the latest Comfyui. hopefully it will fix the issues. Let me know if you have any other questions.

This is a problem with comfyui. If you revert to the old version, the problem will be gone. Go to the comfyui directory and run "git reset --hard cc9cf6d1bd957d764ad418258b61d7e08187573b"

@@murattuanaahmet Hello! Thanks for the recommendation, in my case the problem was in the python version, after updating it everything worked, but I couldn't achieve such similarity as the video, in principle it works at the level of Stable Diffusion Insta Swap or Reactor.

We were able to fix this error using this: github.com/comfyanonymous/ComfyUI/issues/5601 just go to your E:\pinokio\api\comfyui.git\app\comfy\ldm\flux folder and edit the model file with notepad+ and search for "out = self.forward_orig" line 184 and remove ", transformer_options" at the end of that line and save changes this will fix this errors

Make sure all the models are correctly installed

You need to setup an LLM API for it to work, either do it free by installing Ollama see our other video in the description on how to install it, after that watch this video to see how to add the API.

You have to delete the mflux app completely to delete the models as they are hidden.