Hey there! great tut, what's the difference between text to image and image to image using the method taught? would image to image allow a reference image ?
Hello. I followed the video and configured the same model and Lora, but when I set the Denoising strength to 1, no matter how the Starting Control Step is set, there will be no QR code. What's going on? Looking forward to your answer. 😭
Here is my process, but the generated image does not display any signs of a QR code: head of a robot Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 3600572913, Size: 512x512, Model hash: 4199bcdd14, Model: revAnimated_v122, Denoising strength: 1, Clip skip: 2, ControlNet 0: "preprocessor: tile_resample, model: control_v11f1e_sd15_tile [a371b31b], weight: 1, starting/ending: (0.2, 1), resize mode: Crop and Resize, pixel perfect: True, control mode: Balanced, preprocessor params: (-1, 1, -1)"
Your settings look good to me. Do you get nothing at all when trying to generate an image? You should at least see one image of the same QR code that you use as input (which is the result of the controlnet preprocessing). I’ve had a similar issue once where out of a sudden no QR codes were generated anymore. Restarting Automatic1111 didn’t help either and what fixed it was restarting my pc. If you tried that already and still not getting any results, check the command window and see if there are errors there.
I’ve had that issue a couple of times before upgrading my GPU. I think it mainly occurred when trying to generate images in a high resolution (or in a non “standard“ resolution that wasn’t 512x512. You could try editing the webui-User.bat by setting “set COMMANDLINE_ARGS=--medvram --opt-split-attention“. And make sure you have enough free disk space. A clean install of automatic1111 could help too if especially if you’re using an older version. There’s some threads on git which might help: github.com/AUTOMATIC1111/stable-diffusion-webui/issues/3527 and github.com/AUTOMATIC1111/stable-diffusion-webui/issues/302 . Hope you’ll be able to figure it out
The only thing you need (that isn't explained in the video) is Automatic1111. Installing is quite easy: github.com/AUTOMATIC1111/stable-diffusion-webui
Great video! I was following along until 4:43 because AttributeError: 'Options' object has no attribute 'lora_functional' popped up and it wasn't showing the 2 QR codes after I clicked Generate. Can you help me solve this issue? TIA!
Thanks! I found a few threads on git which describe an error similar to what you have (github.com/vladmandic/automatic/issues/225). However, there's no clear answer how to fix it. Are you using an extension for the LoRAs? Seems this extension is causing issues and is only meant for WebUIs that do not support the LoRAs out of the box. You might have to switch the WebUI to A1111 if you're not using it already. Otherwise I'd try to update A1111. Hope this helps
In Preprocessor tile_resample ok! but when I go to Model there is nothing in the label, How do I put controlnet11Models_tileE In Model? Can you help me please?
@@AiVOICETUTOR Thank you I managed to do following all the steps you taught thank you very much. You have gained a loyal follower. ruclips.net/user/shortsfxsHnbugCS8
Sorry if it wasn't clear enough. To install the model you only need to download it (the .safetensors file) and put it in stable-diffusion-webui/models/Lora. Then you can use the Lora model by putting the activation and trigger words into the prompt.
I‘m using 5a0f8864. I’ve had a similar issue once. Try changing the value of the max model amount of MultiControlNet to something other than what it is currently set to (and larger larger than 1). You find it under stable diffusion settings ->controlnet. Then restart the entire webUI
The QR scanning on Android varies a lot. For example Samsung phones are said to be very bad at it compared to others. So its better to use a third party app on Android
![Quando Rondo - Life Goes On [Official Music Video]](http://i.ytimg.com/vi/eIf3fhMXAow/mqdefault.jpg)
Hey there! great tut, what's the difference between text to image and image to image using the method taught? would image to image allow a reference image ?
Thanks for the video! It was easy to follow and create awesome QR codes.
Awesome. I’m glad you liked it.
Hello. I followed the video and configured the same model and Lora, but when I set the Denoising strength to 1, no matter how the Starting Control Step is set, there will be no QR code. What's going on? Looking forward to your answer. 😭
Replied to your other comment. Hope you’ll be able to get it to work
Your guide was easy and good.
it was so much fun from korea
Thank you very much! Greetings to Korea
Already installed ControlNet the same way as you did.
But why it not show "ControlNet Unit 0", "ControlNet Unit 1". It allow upload 1 image only.
Try going into the controlnet settings and change the number of controlnets
@@AiVOICETUTOR works now, thanks for helping.
Awesome. You're welcome!
Here is my process, but the generated image does not display any signs of a QR code:
head of a robot
Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 3600572913, Size: 512x512, Model hash: 4199bcdd14, Model: revAnimated_v122, Denoising strength: 1, Clip skip: 2, ControlNet 0: "preprocessor: tile_resample, model: control_v11f1e_sd15_tile [a371b31b], weight: 1, starting/ending: (0.2, 1), resize mode: Crop and Resize, pixel perfect: True, control mode: Balanced, preprocessor params: (-1, 1, -1)"
Your settings look good to me. Do you get nothing at all when trying to generate an image? You should at least see one image of the same QR code that you use as input (which is the result of the controlnet preprocessing). I’ve had a similar issue once where out of a sudden no QR codes were generated anymore. Restarting Automatic1111 didn’t help either and what fixed it was restarting my pc. If you tried that already and still not getting any results, check the command window and see if there are errors there.
FYI, you don't need to put the same image in your first control net, if you're using the save image, control net knows if you enable control net 🙂🙂
Thanks for letting me know!
@@AiVOICETUTOR why is the ai image generation community so wholesome? 😊
@@AiVOICETUTOR Does this mean that another image can be inserted to use as reference?
Thank you for this video! I am getting an error when I try to generate with this model that my CUDA is out of memory. Is this a common issue?
I’ve had that issue a couple of times before upgrading my GPU. I think it mainly occurred when trying to generate images in a high resolution (or in a non “standard“ resolution that wasn’t 512x512. You could try editing the webui-User.bat by setting “set COMMANDLINE_ARGS=--medvram --opt-split-attention“. And make sure you have enough free disk space. A clean install of automatic1111 could help too if especially if you’re using an older version. There’s some threads on git which might help: github.com/AUTOMATIC1111/stable-diffusion-webui/issues/3527 and github.com/AUTOMATIC1111/stable-diffusion-webui/issues/302 . Hope you’ll be able to figure it out
How do you use stable diffusion in brave ?
You can open the URL from Automatic1111 in any browser. There's also ways to run it completely in the cloud (Colab or ThinkDiffusion for example)
@@AiVOICETUTOR how do I run it on colab ?
Here is some info on Colab: github.com/camenduru/stable-diffusion-webui-colab. I haven't tried it myself
excellent work
Thank you! Cheers!
Where can I find all the things I need to download all this?
The only thing you need (that isn't explained in the video) is Automatic1111. Installing is quite easy: github.com/AUTOMATIC1111/stable-diffusion-webui
can you prvoide the whole settings? i didn't get good result
Sure I always put all prompts in the video description: pastebin.com/cFtF4Hcu. Hope you’ll manage to get some good results with it
@@AiVOICETUTOR thanks but I mean the setting for the controlnet
Oh gotcha! Sorry it took a while but I added some examples with the full settings: pastebin.com/ZyHZKayF
Great video! I was following along until 4:43 because AttributeError: 'Options' object has no attribute 'lora_functional' popped up and it wasn't showing the 2 QR codes after I clicked Generate. Can you help me solve this issue? TIA!
Thanks! I found a few threads on git which describe an error similar to what you have (github.com/vladmandic/automatic/issues/225). However, there's no clear answer how to fix it. Are you using an extension for the LoRAs? Seems this extension is causing issues and is only meant for WebUIs that do not support the LoRAs out of the box. You might have to switch the WebUI to A1111 if you're not using it already. Otherwise I'd try to update A1111. Hope this helps
thank you:)
You're welcome!
In Preprocessor tile_resample ok! but when I go to Model there is nothing in the label, How do I put controlnet11Models_tileE In Model? Can you help me please?
Download the model from huggingface.co/lllyasviel/ControlNet-v1-1/tree/main and put it in stable-diffusion-webui/models/ControlNet
@@AiVOICETUTOR thank you so much😘❤️❤️❤️
Excellent tutorial I subscribed your channel😉
Thanks! Much appreciated ❤️
@@AiVOICETUTOR Thank you I managed to do following all the steps you taught thank you very much. You have gained a loyal follower. ruclips.net/user/shortsfxsHnbugCS8
Didn't understand the Install Lora Models Part
Sorry if it wasn't clear enough. To install the model you only need to download it (the .safetensors file) and put it in stable-diffusion-webui/models/Lora. Then you can use the Lora model by putting the activation and trigger words into the prompt.
@@AiVOICETUTOR Thanks! Also which path file do I need to download?
If you mean the .PTH file for Controlnet, then you only need one. Its mentioned here: ruclips.net/video/Y1rYTnupZk4/видео.html
which controlnet web ui version are you using? mine doesn't show second controlnet tab
I‘m using 5a0f8864. I’ve had a similar issue once. Try changing the value of the max model amount of MultiControlNet to something other than what it is currently set to (and larger larger than 1). You find it under stable diffusion settings ->controlnet. Then restart the entire webUI
are they scannable?
Yes they all scan on iOS. Some more info about scanning in my other video: ruclips.net/video/nFfHEmNu0Zk/видео.html
@@AiVOICETUTOR what about on Android phones?
The QR scanning on Android varies a lot. For example Samsung phones are said to be very bad at it compared to others. So its better to use a third party app on Android
i could sold that for business and peaple?
In theory you could do so, but I'd be careful using it in a commercial context, since the question about copyright isn't fully answered yet.
Create your images with any ai tools…I can make an scannable QR Code 😀
Hope you're having fun with it 😃
3:06
this process is just ridiculous LOL
😇