Software Engineering Courses (SECourses)•3y ago

hello, I was watching your latest tutorial: Double Your Stable Diffusion Inference Speed with RTX Ac

hello, I was watching your latest tutorial: Double Your Stable Diffusion Inference Speed with RTX Acceleration TensorRT: A Comprehensive Guide, & I was wondering does all these settings apply to RunPod SD as well?
can I dl the Nvidia extension for my network volume on runpod?

David•12/11/23, 12:17 AM

Hi! Anybody knows here how https://www.instagram.com/p/Cs7ECvghZYF/?igshid=MzRlODBiNWFlZA== is able to have zuch realism. I have myself super good result with skin textures and hair but this one is to me the best out there.

soymarcelojs•12/11/23, 12:20 AM

hello! I'm new to the group! I am doing the optimizations of the SD installation but I get this error when I install them until the xformers update part

soymarcelojs•12/11/23, 12:21 AM

Does anyone know how I can solve this error?

Pelolisu•12/11/23, 12:21 AM

NNed Shoaei hello, I was watching your latest tutorial: Double Your Stable Diffusion Inferen...

Furkan Gözükara SECourses•12/11/23, 12:27 AM

yes works on runpod too

Furkan Gözükara SECourses•12/11/23, 12:27 AM

https://youtu.be/eKnMVXVjVoU

YouTubeSECourses

How to Install & Run TensorRT on RunPod, Unix, Linux for 2x Faster ...

Stable Diffusion Gets A Major Boost With RTX Acceleration. One of the most common ways to use Stable Diffusion, the popular Generative AI tool that allows users to produce images from simple text descriptions, is through the Stable Diffusion Web UI by Automatic1111. In today’s Game Ready Driver, NVIDIA added TensorRT acceleration for Stable Diff...

Ned ShoaeiOP•12/11/23, 12:28 AM

thank you! Nvidia is breaking grounds in SD!

Ssoymarcelojs hello! I'm new to the group! I am doing the optimizations of the SD installation...

Furkan Gözükara SECourses•12/11/23, 12:28 AM

disable xformers and generate TensorRT

FFurkan Gözükara SECourses you can get many times better results than this with SDXL DreamBooth

Pelolisu•12/11/23, 12:28 AM

i can run in kaggle

PPelolisu i can run in kaggle

Furkan Gözükara SECourses•12/11/23, 12:28 AM

yes Kaggle also has SDXL

Furkan Gözükara SECourses•12/11/23, 12:28 AM

but not as good as on PC or RunPod

Furkan Gözükara SECourses•12/11/23, 12:28 AM

due to FP16 limitation

Pelolisu•12/11/23, 12:29 AM

on pc i can run on Bf16

PPelolisu on pc i can run on Bf16

Furkan Gözükara SECourses•12/11/23, 12:29 AM

what is your vram

Pelolisu•12/11/23, 12:29 AM

8gb + 16gb shared

Pelolisu•12/11/23, 12:29 AM

3060 ti

Furkan Gözükara SECourses•12/11/23, 12:29 AM

ah you have too bad gpu

Furkan Gözükara SECourses•12/11/23, 12:30 AM

shared will take forever

Pelolisu•12/11/23, 12:31 AM

I have a doubt. Does the vram impact the training result or just the training speed?

Pelolisu•12/11/23, 12:37 AM

However, my Drembooth on the A1111 does not exceed 7.5gb of use even though I do not specify --medvram, perhaps the algorithm is adjusting itself.

KGTachi•12/11/23, 12:38 AM

@Dr. Furkan Gözükara hi Dr which version is the best to fresh install stable diffusion from your patreon with latest xformers

FFurkan Gözükara SECourses disable xformers and generate TensorRT

soymarcelojs•12/11/23, 12:42 AM

now i have this problem! sorry im noob

Pelolisu•12/11/23, 12:46 AM

i hate this errors lol so i moved my env to wsl 2 and fixi ng it one by one. On Windows this erros can ^3

Pelolisu•12/11/23, 12:47 AM

The problem with this is that each case is different and depends a lot on the combinations of dependency versions

soymarcelojs•12/11/23, 12:52 AM

Well, I'll see how I can solve it. Take a step at least I have a tip

Ssoymarcelojs now i have this problem! sorry im noob

Furkan Gözükara SECourses•12/11/23, 12:53 AM

you have multiple issues

Furkan Gözükara SECourses•12/11/23, 12:53 AM

i would install python into C drive

Furkan Gözükara SECourses•12/11/23, 12:53 AM

without any empty spaces in path

Furkan Gözükara SECourses•12/11/23, 12:53 AM

i would do same with automatic1111 too

Furkan Gözükara SECourses•12/11/23, 12:53 AM

install into any drive without any space letters

Furkan Gözükara SECourses•12/11/23, 12:54 AM

in this video i have shown how to install automatic1111 and python properly

Furkan Gözükara SECourses•12/11/23, 12:54 AM

https://youtu.be/B5U7LJOvH6g

YouTubeSECourses

How To Install Python, Setup Virtual Environment VENV, Set Default ...

How to install Python, have multiple Python installations, set system wide default Python version. How to compose venv for any Python installation, change Python default path, and install SD web UI properly. Discord: https://bit.ly/SECoursesDiscord.

If I have been of assistance to you and you would like to show your support for my work, please...

Furkan Gözükara SECourses•12/11/23, 12:54 AM

also this video : https://youtu.be/kvxX6NrPtEk

YouTubeSECourses

Double Your Stable Diffusion Inference Speed with RTX Acceleration ...

Furkan Gözükara SECourses•12/11/23, 12:54 AM

you have 3 space letters having folders

Furkan Gözükara SECourses•12/11/23, 12:54 AM

big no no

Furkan Gözükara SECourses•12/11/23, 12:59 AM

also your cmd doesn't give any info about the error

soymarcelojs•12/11/23, 12:59 AM

thanks!

soymarcelojs•12/11/23, 12:59 AM

ill see that now!

Digital [Starburst]•12/11/23, 1:02 AM

The RunPod setup tutorial is a bit outdated because they have updated the platform. You no longer have a choice between Stable-Diffusion-1.5 and Stable-Diffusion-2.1. You now must choose between Fast-Stable-Diffusion, ComfyUI-Stable-Diffusion, Stable-Diffusion-webui, and I think there's one other option. Under the Community options the A5000 is still an option. There is one other option for the same price that has more CPU RAM, but I don't know if you want that one.

soymarcelojs•12/11/23, 1:16 AM

I thank you, but I'm not there yet. I'm trying to get stability to run SDWebUI. Do you suggest I watch that class? because I was stuck in the second one even for this issue

FFurkan Gözükara SECourses also this video : https://youtu.be/kvxX6NrPtEk

Lorelai AI•12/11/23, 1:21 AM

Can the TensorRT do images above 1024X1024? I need images by 768x1344 and I don't think I could get it to work when I used to use it because sizes are capped?

Pelolisu•12/11/23, 1:30 AM

@Dr. Furkan Gözükara have you managed to get Drembooth to work locally with DeepSpeed? If so, have you had good results with that?

Jehex•12/11/23, 2:10 AM

Hi, again an error of path right ?

Hustle with AI•12/11/23, 4:22 AM

Hello, very new to SD. I have installed Dreambooth extension in web UI using auto installer scripts on Runpod. I am trying to create what's shown in Corridor Digital crew video.

Any help to use Dreambooth like they did in using 4 datasets to train a model?

Dataset1: Niko - The main subject (a real man's images for training data set)
Dataset2: Ghibli Animation characters - images to train the main subject on
Dataset3: General images of man (reg images)
Dataset4: General images of art style (reg images)

Benjamin•12/11/23, 10:26 AM

@FiveSAI sorry if I missed it, but did you already share your training parameters for sdxl dreambooth?

interested in trying them out

Ssoymarcelojs I thank you, but I'm not there yet. I'm trying to get stability to run SDWebUI. ...

Furkan Gözükara SECourses•12/11/23, 3:01 PM

i got up to date tutorials and install scripts

Furkan Gözükara SECourses•12/11/23, 3:01 PM

what you need

LLorelai AI Can the TensorRT do images above 1024X1024? I need images by 768x1344 and I don'...

Furkan Gözükara SECourses•12/11/23, 3:01 PM

yes it can do

Furkan Gözükara SECourses•12/11/23, 3:01 PM

you need to modify 1 file though

Furkan Gözükara SECourses•12/11/23, 3:02 PM

https://github.com/NVIDIA/Stable-Diffusion-WebUI-TensorRT/issues/199

GitHub

Maximum resolution is limited to 1024 but sliders goes up to 2048 h...

I had to modify this code to generate TensorRT that will support highres fix for SDXL self.max_image_shape = 768 if version in ("1.4", "1.5") else 2048 It is by default 1024 So ...

hello, I was watching your latest tutorial: Double Your Stable Diffusion Inference Speed with RTX Ac

Similar Threads

Similar Threads

Similar Threads