Finally got it working. If you download directly from hugging face make sure you create a Stable-Cascade folder inside the unnet folder and then place the files. Else you will get error.
It seems the website that I used for sharing the workflows has some issues. I have updated the links in the description with working links. Thanks for letting me know. It should work fine now.
hi can you help me to solve this error "module diffuser has no attribute StableCascadeUnet. i installed cascade install in stable diffusion but i got this error after installing of all model on window 11
That pop-up in the bottom right of your screen at 3:15, if that's a recurring issue it's because you chose UK English over US English on install. I've had this issue 🤣
Is that so? I always wondered why it keeps popping up. Haha. I thought it was a normal feature of the windows. I'll try changing the system language to US english to see if this stops popping up.
Yep and it'll fix a couple of keyboard quirks too lol, that particular installation can't do anything but US. Play around with language settings and you'll eventually fix it, it gave me such a headache haha, good luck 😁
To be honest I did not expect a 1050TI to run cascade at all. That's great. When I had 4GB 1650, I remember it crashing quite a lot with SD 1.5 And it took so long to generate, I used to get quite frustrated with that.
@@Debargha_Bhattacharjee Wow really? I also run 1.5 and sdxl, along with a lot of control nets and area composition, never had a crash; but yes, some can take a long time. Do you think low Vram effect the quality of the pic? Or just speed...
V-ram does not affect the quality of the image at all. It just makes image generation slower or faster. But if the amount of vram becomes too low for the requirements, it can generate broken images or mostly black screens. I had quite a few black screen issues back in the day with custom checkpoints and gtx1650 whenever I tried to generate images larger than 512x512.
La primera imagen me creó perfecta del pingüino y luego las otras imágenes no logra crearlos queda como una especie de barro que se le debe? Tengo una tarjeta nvidia 3070 8vram
Have you tried restarting the comfyui or launching the workflow again? This might solve the issue. if not, then try to update the ui and the extensions, that should solve it.
Just load the given workflow into comfyui. Then go to 'missing custom nodes'. There it would show the necessary nodes to be installed for this to work. Install them and restart the UI. It should start showing up.
![[UNCUT] “บอสณวัฒน์” ลากไส้!! ใครหนุนหลัง "แม่ตั๊ก ป๋าเบียร์" I คนดังนั่งเคลียร์ I 30 ก.ย. 67](http://i.ytimg.com/vi/QlD7mU2KKRk/mqdefault.jpg)
Hi bro., are there any ability to add upscale to it? can it also be used with control net? or ipadapter?
I shall try looking into it. There should definitely be a way. I'll try some nodes to see what works out.
Stable diffusion and stable cascade is two different things plz someone explain
Thank you! I will be testing these today. Very informative.
Finally got it working. If you download directly from hugging face make sure you create a Stable-Cascade folder inside the unnet folder and then place the files. Else you will get error.
Thanks a lot for the info. That's really helpful.
great video straight to the point and gets me to where i need to go. Great work!
Thank you. I'm glad it was helpful.
Stable Cascade with Fooocus Styler Workflow show 404 error Sorry, page not found.
It seems the website that I used for sharing the workflows has some issues. I have updated the links in the description with working links. Thanks for letting me know. It should work fine now.
@@Debargha_Bhattacharjee it's broke again
hi can you help me to solve this error "module diffuser has no attribute StableCascadeUnet. i installed cascade install in stable diffusion but i got this error after installing of all model on window 11
fyi my 4060 16GB VRAM executed (on the second run after the models had been loaded into RAM) in 19.74 seconds using bf16 models
Thanks for letting me know. Looks like it's really going to be a real gamechanger.
@@Debargha_BhattacharjeeOn my 8GB 4060 it's only slightly slower, 25.49 seconds with bf16.
Edit: also the workflow links in the description are dead.
I am really grateful that you let me know about the issue. Just updated the workflow link with another website. Kindly check if it's working.
No you don't i confirm, on the rtx 4090 without any special setup on comfyUI you reach the 12.5 GB on the Vram on a 4 img vision workflow
Yeah, I was overjoyed to learn that. Earlier I was heartbroken thinking that I might never get to run this on my pc.
That pop-up in the bottom right of your screen at 3:15, if that's a recurring issue it's because you chose UK English over US English on install. I've had this issue 🤣
Is that so? I always wondered why it keeps popping up. Haha. I thought it was a normal feature of the windows. I'll try changing the system language to US english to see if this stops popping up.
Yep and it'll fix a couple of keyboard quirks too lol, that particular installation can't do anything but US. Play around with language settings and you'll eventually fix it, it gave me such a headache haha, good luck 😁
My crap ass 4G 1050TI runs cascades' largest model. takes a few mins... but runs.
To be honest I did not expect a 1050TI to run cascade at all. That's great. When I had 4GB 1650, I remember it crashing quite a lot with SD 1.5 And it took so long to generate, I used to get quite frustrated with that.
@@Debargha_Bhattacharjee Wow really? I also run 1.5 and sdxl, along with a lot of control nets and area composition, never had a crash; but yes, some can take a long time.
Do you think low Vram effect the quality of the pic? Or just speed...
V-ram does not affect the quality of the image at all. It just makes image generation slower or faster. But if the amount of vram becomes too low for the requirements, it can generate broken images or mostly black screens. I had quite a few black screen issues back in the day with custom checkpoints and gtx1650 whenever I tried to generate images larger than 512x512.
La primera imagen me creó perfecta del pingüino y luego las otras imágenes no logra crearlos queda como una especie de barro que se le debe? Tengo una tarjeta nvidia 3070 8vram
Have you tried restarting the comfyui or launching the workflow again? This might solve the issue. if not, then try to update the ui and the extensions, that should solve it.
where to get foocus style selector as its not coming in comfyui manager
Just load the given workflow into comfyui. Then go to 'missing custom nodes'. There it would show the necessary nodes to be installed for this to work. Install them and restart the UI. It should start showing up.
wow, your pc spec is exactly same as mine, from ram, gpu, cpu and even the monitor 😮
That's a nice setup you got there.
@@Debargha_Bhattacharjee thanks 😁