Open FurkanGozukara opened 9 months ago
Just as slow as pinokio in my case.... maybe a tiny bit faster. Slightly annoyed. 4070 8gb vram on low gave me upwards of a 7 min gen.
Just as slow as pinokio in my case.... maybe a tiny bit faster. Slightly annoyed. 4070 8gb vram on low gave me upwards of a 7 min gen.
My Patreon follower generate images with 2+it second with RTX 4070 mobile. So it would take roughly 1 min. So you are doing mistaken comparison. My app is not pinokio app. I developed this. Supports CPU offloading
Just as slow as pinokio in my case.... maybe a tiny bit faster. Slightly annoyed. 4070 8gb vram on low gave me upwards of a 7 min gen.
My Patreon follower generate images with 2+it second with RTX 4070 mobile. So it would take roughly 1 min. So you are doing mistaken comparison. My app is not pinokio app. I developed this. Supports CPU offloading
I'm a Patreon of yours and I'm not getting that. I'm sorry. I have 32 non vram and I'm not computer savvy obviously. So I may be doing something wrong, but these are taking a good while. Is there anything known to slow them down?
Just as slow as pinokio in my case.... maybe a tiny bit faster. Slightly annoyed. 4070 8gb vram on low gave me upwards of a 7 min gen.
My Patreon follower generate images with 2+it second with RTX 4070 mobile. So it would take roughly 1 min. So you are doing mistaken comparison. My app is not pinokio app. I developed this. Supports CPU offloading
I'm a Patreon of yours and I'm not getting that. I'm sorry. I have 32 non vram and I'm not computer savvy obviously. So I may be doing something wrong, but these are taking a good while. Is there anything known to slow them down?
can you reinstall latest version and also start with this option and let me know? before starting that make sure that you minimized your GPU vram usage
if you cant still solve message me from Patreon i can take deeper look
This ad for a paywalled system should be deleted.
This ad for a paywalled system should be deleted.
This is basically ads of this model itself
Posts in issues, titles it not an issue. Fuckin ad, I see you all over various issue boards of various models and im so damn tired of you shilling your paywalled content or youtube videos. Hopefully mods delete this trash to allow people to see actual issues and potential fixes instead of getting blasted in the face by an ad
Posts in issues, titles it not an issue.
Fuckin ad, I see you all over various issue boards of various models and im so damn tired of you shilling your paywalled content or youtube videos.
Hopefully mods delete this trash to allow people to see actual issues and potential fixes instead of getting blasted in the face by an ad
what's your problem? people may actually not be able to cascade test without diffusers knowledge. If you don't like this help why even open the non-issue thread. Realax man
Posts in issues, titles it not an issue. Fuckin ad, I see you all over various issue boards of various models and im so damn tired of you shilling your paywalled content or youtube videos. Hopefully mods delete this trash to allow people to see actual issues and potential fixes instead of getting blasted in the face by an ad
what's your problem? people may actually not be able to cascade test without diffusers knowledge. If you don't like this help why even open the non-issue thread. Realax man
The issue is that git is for openly sharing source code. Not for shilling paywalled systems. You don't see other devs using git issues as an advertising platform. It just comes across as "Hey, look what I did with these models! You want to try them? Ha ha, sorry, pay me first!" I could easily put ads all over git for my software, but I (and 99% of other devs thankfully) do not. That is why this annoys people.
Posts in issues, titles it not an issue. Fuckin ad, I see you all over various issue boards of various models and im so damn tired of you shilling your paywalled content or youtube videos. Hopefully mods delete this trash to allow people to see actual issues and potential fixes instead of getting blasted in the face by an ad
what's your problem? people may actually not be able to cascade test without diffusers knowledge. If you don't like this help why even open the non-issue thread. Realax man
This dude literally posts a paywalled link in a thread asking for help, then posts it here as an ad, and has in the past posted shit like this everywhere else. This isnt 'here is some code to help your issues' its 'pay me to get this code.' Thats like... the antithesis of 'open source' is it not? If it was 'here is code please donate if you like it' there would be 0 issues but its not. This isnt him doing stuff for the good of everyone else, its just to get money.
while you keep complaining about me i am reporting so many problems and getting them fixed
https://github.com/huggingface/diffusers/pull/6487#issuecomment-1944380496
https://github.com/huggingface/diffusers/pull/6487#issuecomment-1944322796
https://github.com/huggingface/diffusers/pull/6487#issuecomment-1944837287
https://github.com/gradio-app/gradio/issues/7391#issuecomment-1939710069
if you want to help the community do something else
while you keep complaining about me i am reporting so many problems and getting them fixed
The complaints have nothing to do with your huggingface posts. They are about you posting your ad here. You really do not understand? Your OP post here is helping nobody and not fixing any problems. It is just an ad trying to get people to pay money to join your paywalled Patreon. It does not belong on github.
We added FP16 support for older GPUs such as rtx 1080 TI and also Kaggle notebook with gradio for free Kaggle tier that provides dual T4 gpu. FP16 support was mandatory for Kaggle since those GPUs don't support BF16. Both working great right now.
You can download our scripts here : https://www.patreon.com/posts/98410661
I have developed an amazing 1 click installer and Gradio app for this newest amazing model.
This app working many times faster than the Jupyter Notebook shared here. It uses Diffusers Library.
Auto download everything including models
This model can be considered as next level of Stable Diffusion.
The gradio APP I developed supports low VRAM and works great on even 8 GB GPUs
Saves every generated image automatically in outputs folder and many a lot of improvements
Kaggle not working right now due to FP16 bug and I have reported it to be fixed. Hopefully after that notebook will work great
Batch size 4, 1536x1280 resolution it / s is 1.7 on RTX 4090
Batch size 1, 1024x1024 resolution it / s is 12.14 (encoder) / 10.6 (decoder) on RTX 4090
So 1 image takes like 4 seconds on RTX 4090 for 1024x1024