Stable Diffusion AI: Options to run using CPU-only to Low GPU specs, Android, etc.

[alist1986]

From latest Stable Diffusion development, it was stated You do not have permission to view the full content of this post. Log in or register now. that it will acceptably run on Windows 10 or higher OS with a discrete Nvidia video card (GPU) with a 4 GB VRAM or more - the minimum graphic hardware specification for You do not have permission to view the full content of this post. Log in or register now.. Meaning, the GPU will run/process the application - not the cpu by default. If it does, you'll ran out of memory and experience pc failures. If you look at the SD GPU performance benchmarks You do not have permission to view the full content of this post. Log in or register now., you'll see the big difference on how the top GPUs perform using SD, what more if it is none on the list. For those of us who don't own yet the required hardware (like me), there are still ways to install and try regardless of how long it process your images.
People can still use the opensource Stable Diffusion using free trial demo online or simply using the notebooks for a limited time via Google Colab service. It;s the best option.
But I'll focus on the unlimited offline use of some SD forks which work. I'm not an expert on Stable Diffusion usage, so I'm mainly concentrating on possible non-standard methods that work with only cpu or low end GPU at best. So bear with me.

First the first optionwe will use: You do not have permission to view the full content of this post. Log in or register now. . A lightweight SD version similar to You do not have permission to view the full content of this post. Log in or register now. - another SD-cpuonly option.

Stable Diffusion CPU only​

This fork of Stable-Diffusion doesn't require a high end graphics card and runs exclusively on your cpu. It's been tested on Linux Mint 22.04 and Windows 10.

This isn't the fastest experience you'll have with stable diffusion but it does allow you to use it and most of the current set of features floating around on the internet such as txt2img, img2img, image upscaling with Real-ESRGAN and better faces with GFPGAN.

A install guide video is located here. You do not have permission to view the full content of this post. Log in or register now.

​

Requirements​

Windows and Linux requirements​

Install Anaconda You do not have permission to view the full content of this post. Log in or register now. Yes even on a linux system anaconda needs to be installed.

Windows​

Install Visual Studio Community Edition to build 1 file.​

Click on the free download and make sure to check "Desktop Development with C++" when installing as per this image (You do not have permission to view the full content of this post. Log in or register now.)

You do not have permission to view the full content of this post. Log in or register now.

Install Git​

You do not have permission to view the full content of this post. Log in or register now. version control manager for code we just use it to download repos from GitHub Must be on system PATH, When installing select the option add to system PATH

Install Wget​

used to download models for projects Windows users need this verison You do not have permission to view the full content of this post. Log in or register now. download the .exe and I copied it to my C:/Windows/System directory (this isn't the correct way just the fastest to get it working)

reboot your system just to make sure everything is properly loaded up.

Linux Mint 22.04​

Install git and wget with the following command

sudo apt-get -y install git wget build-essential

Installation of Stable-Diffusion-cpuonly​

Download Stable-Diffusion-cpuonly​

Copy this github repository and extract the files.

Download the CompVis Stable-diffusion model.​

Go here and download the correct mode from here. You'll have to agree to the license setup an account but this is the bread and butter AI art generating learning model.

You do not have permission to view the full content of this post. Log in or register now.

copy the file to your stable-diffusion-cpuonly-main directory

Download the GFPGAN model​

You do not have permission to view the full content of this post. Log in or register now.

copy the file to your stable-diffusion-cpuponly-main directory

Windows - Running the install script​

open a terminal or powershell and cd to your stable-diffusion-cpuonly-main directory and run

.\install_sdco.bat

Linux - Running the install script​

open a terminal or powershell and cd to your stable-diffusion-cpuonly-main directory and run

bash -i install_sdco.sh

Windows - Starting Stable-Diffusion-cpuonly​

Run the following command

.\run_sdco.bat

Linux - Starting Stable-Diffusion-cpuonly​

Run the following command

bash -i run_sdco.sh

This is the fork and guide I initially used for the trial but you need to replace line#17 of install_sdco.bat with this code:

call pip install -e git+https://github.com/crowsonkb/k-diffusion#egg=k_diffusion

Previous github link was abandoned so you'll end up with a missing module.
I already mentioned this on another related You do not have permission to view the full content of this post. Log in or register now.. Here is the pic.

My 2nd try using the same test prompts resulted in much better graphics compared to the link above. Though SD-cpuonly works, it took me ~900 sec to create an image. I'm not satisfied, but that is the result. The only consolation is the use of GFPGAN (a good image restorer) plus Real-ESGRAN, a high end upscaler and some other standard image processing tools.
BTW, I'm using an Intel Core i7 3632QM with 12GB of RAM (with 2GB NVIDIA GT 740m with CUDA 10 drivers but not used). It took around +8GB to run the app and open the GUI via browser link (You do not have permission to view the full content of this post. Log in or register now.). After that, it took +4GB of memory to process/create the image at that given timeframe.

You can also try using You do not have permission to view the full content of this post. Log in or register now. from this link, place it on your working directory and restart install_sdco.bat again to replace your model.ckpt. Then see the results on your next render.
BTW, there is also a Web GUI for this by the same forker: You do not have permission to view the full content of this post. Log in or register now.
Check the new features if you want to try.
Note: Please read the whole development page to know the features and limitations. At the moment, it can't use controlnet nor safetensors. You can try merging models using You do not have permission to view the full content of this post. Log in or register now. (via Google Colab) or try the tools provided here: You do not have permission to view the full content of this post. Log in or register now.. Do it at your own risk and create your hybrids he he. Use your imagination.
Thanks. I'll update later.

PS: Check this thread "You do not have permission to view the full content of this post. Log in or register now." and "You do not have permission to view the full content of this post. Log in or register now." to try SD, or make an "You do not have permission to view the full content of this post. Log in or register now.".

 

[alist1986]

2ND OPTION: Worked fine with more featured/user-friendly WEBUI, but no option yet to use safetensors, controlnet,...A1111 is still advanced comparing to this. (But we're only interested in using it with CPU.)

Easy Diffusion: AIO installer - You do not have permission to view the full content of this post. Log in or register now.​

Easy Diffusion is an easy to install and use distribution of You do not have permission to view the full content of this post. Log in or register now., the leading open source text-to-image AI software. Easy Diffusion installs all required software components required to run Stable Diffusion plus its own user friendly and powerful web interface for free.

The easiest way to install and use You do not have permission to view the full content of this post. Log in or register now. on your computer.​

Does not require technical knowledge, does not require pre-installed software. 1-click install, powerful features, friendly community.
Download it here and follow the simple guide: You do not have permission to view the full content of this post. Log in or register now.

You can set to use CPU on the WEBUI.

But you need more CPU Ram to use your pc while its rendering images.

 

[Oppenheimer02]

may nakapag try na ba sa low end? katakot eh hahaha

 

[alist1986]

may nakapag try na ba sa low end? katakot eh hahaha

Di ko muna sinubukan sa low-end GPU ito - cpu lang. Kuntento na ako kahit matagal ang processing at di naman masyadong umiinit laptop ko. Yung GPU, kahit data processing, umiinit. Paano pa kung graphics. Luma na kasi. Ayokong pwersahin dahil laptop siya.
Kung binasa mo naman yung specs ng Easy Diffusion, malinaw naman na ang minimum nya ay "Nvidia GPU with 2GB vram" (sa std SD ay 4GB di ba?) bukod pa yung CPU system ram. Kung tutuusin, kahit naman pasok yung GPU requirementa mo ay mag-iinit din yung GPU dahil sa dami ng cores, threads and tasks na gagawin nya. Ito lang ang alam kong problema for normal troubleshooting sa SD using GPU:
You do not have permission to view the full content of this post. Log in or register now.

EASY DIFFUSION
Hardware requirements:

• Windows: NVIDIA graphics card (minimum 2 GB RAM), or run on your CPU.
• Linux: NVIDIA or AMD graphics card (minimum 2 GB RAM), or run on your CPU.
• Mac: M1 or M2, or run on your CPU.
• Minimum 8 GB of system RAM.
• Atleast 25 GB of space on the hard disk.

(Sa Easy Diffusion "Graphics Memory Settings" i-set mo sa "Low". Katumbas yan ng --lowvram sa commandline ng GPU settings. Marami pang pwedeng ipatong like using --no-half, --xformers for speed but decreases quality, --precision-fullor, --precision-autocast, atbp. May settings na pang-cpu at gpu rin. May binabagayan na setting bukod pa doon sa required o kininikilalang commands ng launcher. Marami pang ginagawang pampiga kaya you'll miss the other features. But you still need to know what you're doing. Maraming cases sa net na ok yung Nvidia with 2GB VRAM kahit sa latest SD A1111 tweaking the launching scripts pero, mabagal, lower quality or expect some random errors... )

Kung pasok GB ram ng GPU and/or GB ram ng CPU (ayon sa taas) ay ok ka na sa dalawa. Bihira ang gumagamit ng ganitong mode dahil nga mabagal yung rendering depende sa settings mo at hardware using cpu only or using minimum GPU requirement. Try at your own risk or just be safe using the online services. Pag-isipan mo muna kung worth ba i-try dahil sa online ay libo na ang free services nyan! Ako, software tester na ang hobby ay mag-test lang ng opensource SW to run in non-stard ways noon pa. I always take the risk pero nagbabasa muna ako before the "hands-on".

(Note: Check ninyong maigi yung hyperlinks ko na usually ay kulay blue dahil tinuturo ko doon yung information para tipid dito sa space. Pansin ko kasi pag naka-dark mode, di ko rin makita he he. Paalala ko lang. Thanks.)

Though marami nang forked development ang Stable Diffusion, banggitin ko na lang yung links sa iba. Ito yung mga offline din and free unlimited use. Pero di ko pa tesed yan for cpu or low-end gpu. I'll mention later when successful. check ninyo yung hyperlinks for information and specs. Kalimitan dyan ay 4GB VRAM GPU and requirement since SD ang base app nila. Yung GUI and feature capabilities lang yung nagkaiba. Lahat yan, pwedeng pigain din. Or check their online services via Google Colab or demo links.

You do not have permission to view the full content of this post. Log in or register now.​

(...It is completely Ùn*énsøréd and unfiltered....) similar to You do not have permission to view the full content of this post. Log in or register now.
You do not have permission to view the full content of this post. Log in or register now.
You do not have permission to view the full content of this post. Log in or register now.

You do not have permission to view the full content of this post. Log in or register now.​

You do not have permission to view the full content of this post. Log in or register now.​

 

[biron]

ano po ginagawa nito in one sentence?

 

[alist1986]

Ito yung isa pang tested ko na that also works on cpu mode:
You do not have permission to view the full content of this post. Log in or register now. : A powerful and modular stable diffusion GUI and backend.

Features​

• Nodes/graph/flowchart interface to experiment and create complex Stable Diffusion workflows without needing to code anything.
• Fully supports SD1.x and SD2.x
• Asynchronous Queue system
• Many optimizations: Only re-executes the parts of the workflow that changes between executions.
• Command line option: --lowvram to make it work on GPUs with less than 3GB vram (enabled automatically on GPUs with low vram)
• Works even if you don't have a GPU with: --cpu (slow)
• Can load ckpt, safetensors and diffusers models/checkpoints. Standalone VAEs and CLIP models.
• Embeddings/Textual inversion
• You do not have permission to view the full content of this post. Log in or register now.
• You do not have permission to view the full content of this post. Log in or register now.
• Loading full workflows (with seeds) from generated PNG files.
• Saving/Loading workflows as Json files.
• Nodes interface can be used to create complex workflows like one for You do not have permission to view the full content of this post. Log in or register now. or much more advanced ones.
• You do not have permission to view the full content of this post. Log in or register now.
• You do not have permission to view the full content of this post. Log in or register now. with both regular and inpainting models.
• You do not have permission to view the full content of this post. Log in or register now.
• You do not have permission to view the full content of this post. Log in or register now.
• You do not have permission to view the full content of this post. Log in or register now.
• You do not have permission to view the full content of this post. Log in or register now.
• Latent previews with You do not have permission to view the full content of this post. Log in or register now.
• Starts up very fast.
• Works fully offline: will never download anything.
• You do not have permission to view the full content of this post. Log in or register now. to set the search paths for models.

Just click and check the github link for additional information.
Ang maganda rito, full-feature GUI siya the same as Stable Diffusion Automatic 1111 with AIO installer at pwede gamitin with sa <3GB GPU card at pwede rin mapaandar sa CPU-mode (kahit mabagal). Kung ayaw ninyo, may Colab option din yan. Good for free offline AI Art generation package of your choice.

 

[alist1986]

Sa latest tests ko ng mga Stable Diffusion UIs, ito yung best and stable kaysa mga nauna dito at full feature pa like the original:
You do not have permission to view the full content of this post. Log in or register now.

This is a portable version based on the You do not have permission to view the full content of this post. Log in or register now. version

---

Running on Windows​

1. Download You do not have permission to view the full content of this post. Log in or register now.
2. Unzip the stable-diffusion-portable-main folder anywhere you want
   Root directory preferred, and path shouldn't has spaces and Cyrillic
   Example: D:\stable-diffusion-portable-main
3. Run webui-user-first-run.cmd and wait for a couple seconds
4. When you see the models folder appeared (while cmd working),
   place any model (for example You do not have permission to view the full content of this post. Log in or register now.) in the ...\models\Stable-diffusion directory
   Example of a full path:
   D:\stable-diffusion-portable-main\models\Stable-diffusion\Deliberate_v2.safetensors
5. After complete download your browser will run Stable Diffusion
   The main launcher for the future will be the webui-user.bat file

P.S. Due to Windows specifics, any attempt to block network access may crash the install/update processes, so you will have to rerun .bat again.

Note: To run with cpu only, create a bat file with these scripts:

start webui.bat  --skip-torch-cuda-test --lowvram --precision full --no-half

Name it as "webui-cpu-mode.bat" and execute.
When you see on the console screen "Running on local URL: You do not have permission to view the full content of this post. Log in or register now.", open the link in your browser and try generating art. To stop, press "Control+C", click "Y" and its over.
Enjoy!

 

[alist1986]

Doon po sa nag-iintay ng android support, meron na rin pong dev dito kasama yung iba rin platforms:
You do not have permission to view the full content of this post. Log in or register now.

Stable-Diffusion implemented by You do not have permission to view the full content of this post. Log in or register now. framework based on C++, supported txt2img and img2img!

Zhihu: You do not have permission to view the full content of this post. Log in or register now.

Video: You do not have permission to view the full content of this post. Log in or register now.

txt2img Performance (time pre-it and ram)

per-it	i7-12700 (512x512)	i7-12700 (256x256)	Snapdragon865 (256x256)
slow	4.85s/5.24G(7.07G)	1.05s/3.58G(4.02G)	1.6s/2.2G(2.6G)
fast	2.85s/9.47G(11.29G)	0.65s/5.76G(6.20G)

Check the Readme.md for instructions on your platform
Refer here to get the assets:
You do not have permission to view the full content of this post. Log in or register now.

Ito yung pinakamabilis na option when using cpu using You do not have permission to view the full content of this post. Log in or register now., since SD is designed for GPU to run normally. The other is using Openvino, but I had problems executing the codes at the moment.

Di ko po nasubukan sa android since I don't use it. Sa pc lang. Kaya,
try at your own risk on other platforms.

Hindi rin po ito updated as in You do not have permission to view the full content of this post. Log in or register now. so don't expect the added features. Its good enough to create AI images like the original. Kung gusto ninyong subukan yung full features, just install the original SD A1111 and run it on cpu mode.

 

[MariaGomez]

pwede po ba specs ng laptop ko?

processor: Intel(R) Core(TM) i5-8265U CPU @ 1.60GHz 1.80 GHz Installed ram: 4.00 GB (3.85 GB usable)system type: 64-bit operating system, x64-based processoredition: windows 11 proos build: 22621.1848intel core i5 8th gen nvidia asus laptop FHD 1080 X409FL - FT561T

Intel(R) UHD Graphics 620
NVIDIA GeForce MX250

lods bakit iba UI ng sakin lite version

 

[alist1986]

Una sa lahat, malinaw naman na cpu-only version ang target ng thread ko. So I assumed binasa mo yung requirements ng Stable Diffusion (original) at itong mga versions sa low-specs bago mo ako tinanong. Binigay ko rin yung specs ko sa tests for comparison.
Bawat versions dyan may kanya-kanyang specs na sinabi sa links nila to run with low specs. I'm not a developer (kundi user din) kaya di ako ang makakapagsabi kung aandar yan ng mahusay in your end. Pero bibigyan lang kita ng idea ayon sa aking experience.

Ang minimum requirement nyan to run is this:
For GPU mode :Nvidia video card (GPU) with a 4 GB VRAM (video ram) or more, walang sinabi sa specs ng system ram since GPU naman ang aandar for SD execution pero at least 8GB siguro as standard sa +Win10 OSes ngayon
For CPU mode: + 8GB system ram depende sa version ng SD na gamit mo dahil may versions na malakas kumain at matipid din. Regardless kung potato ang pc mo, yung system ram ang importante, dahil doon ka dadalihin kapag kinapos ayon sa sinapit ng iba. Merong versions na may cpu settings sa GUI at merong wala, pero kailangan mo pang i-force sa cpu-mode sa launcher para mapiga mo ng husto ayon sa sinabi ko sa previous posts. I can't say anything in your case because I don't know your SD version.

Ayon sa specs mo, di ka pwedeng gumamit ng GPU-mode since MX250 2GB siguro gpu nyan. Sa option for cpu-mode ay di ako sure dahil 4GB lang ang system ram mo. Kung umandar yung Win 11 mo, ilan yung natira na pag-open ng Windows, < 2GB? Maswerte ka nga napalitaw mo yung GUI nyan sa initialization. Di ba nag-hang yung Windows mo. Sa processing yari ka na dahil additional +4GB kain nyan depende sa size ng pics at settings na iba.
Yan lang yung masasabi ko sa ngayon.

 

[alist1986]

pwede po ba specs ng laptop ko?

processor: Intel(R) Core(TM) i5-8265U CPU @ 1.60GHz 1.80 GHz Installed ram: 4.00 GB (3.85 GB usable)system type: 64-bit operating system, x64-based processoredition: windows 11 proos build: 22621.1848intel core i5 8th gen nvidia asus laptop FHD 1080 X409FL - FT561T

Intel(R) UHD Graphics 620
NVIDIA GeForce MX250

lods bakit iba UI ng sakin lite version
View attachment 2687401

Kung sakali lang na di mo alam, bisitahin mo itong link to learn more on running SD: You do not have permission to view the full content of this post. Log in or register now.
Hanapin mo dyan yung running in cpu mode para maliwanagan ka pa.
Kumpleto yang guide para sa tulad natin.

Ako limited sa paggamit ng ibang features sa napagana kong original na SDA1111. Di ako makagamit ng malaking ckpt. Hanggang 3GB lang at paisa-isang Lora. At 1 image per session na 5 - 10 minutes pag nilakihan ko above 512x512. Mabagal talaga sa cpu-mode kahit pa yan 12th-gen na i7 cpu. Ang pag-asa na lang ay yang SD-NCNN at SD-Openvino na pinalalakas yung Intel cpus pero bago pa lang na-introduce dyan yung SD kaya nag-iintay pa ako ng updates. Ang pagasa natin ay yung onnx runtime!

Tingnan mo lang yung benchmarks ng gpu vs cpu dito. Magkaroon ka na ng idea kung bakit kailangan mo talaga ng malaking system ram dahil di mabilis ang pytorch sa cpu mode.
You do not have permission to view the full content of this post. Log in or register now.

Ito yung cpu guide doon sa gustong sumubok:
You do not have permission to view the full content of this post. Log in or register now.

 

[MariaGomez]

Magkano po kaya laptop na maganda sa SD? 40k bili ko ng Laptop ko last 2020 di pa pala sapat to

Kung sakali lang na di mo alam, bisitahin mo itong link to learn more on running SD: You do not have permission to view the full content of this post. Log in or register now.
Hanapin mo dyan yung running in cpu mode para maliwanagan ka pa.
Kumpleto yang guide para sa tulad natin.

Ako limited sa paggamit ng ibang features sa napagana kong original na SDA1111. Di ako makagamit ng malaking ckpt. Hanggang 3GB lang at paisa-isang Lora. At 1 image per session na 5 - 10 minutes pag nilakihan ko above 512x512. Mabagal talaga sa cpu-mode kahit pa yan 12th-gen na i7 cpu. Ang pag-asa na lang ay yang SD-NCNN at SD-Openvino na pinalalakas yung Intel cpus pero bago pa lang na-introduce dyan yung SD kaya nag-iintay pa ako ng updates. Ang pagasa natin ay yung onnx runtime!

Tingnan mo lang yung benchmarks ng gpu vs cpu dito. Magkaroon ka na ng idea kung bakit kailangan mo talaga ng malaking system ram dahil di mabilis ang pytorch sa cpu mode.
You do not have permission to view the full content of this post. Log in or register now.

Ito yung cpu guide doon sa gustong sumubok:
You do not have permission to view the full content of this post. Log in or register now.

Gusto ko po sana mainstall tlga sya sa system ng laptop di yun colab lang at yun pwde talaga sya magamit ng sagad

 

[alist1986]

Magkano po kaya laptop na maganda sa SD? 40k bili ko ng Laptop ko last 2020 di pa pala sapat to


Gusto ko po sana mainstall tlga sya sa system ng laptop di yun colab lang at yun pwde talaga sya magamit ng sagad

Kung full feature ng mga top ai art generatorsl like SD, siyempre ang starting point mo yung pinakamura na GPU na supported yung deep neural network using cuda, tensorrt, cudnn dahil doon na punta ng development nyan para may leverage ka sa future updates, which is RTX-2060??? at the least. Pero RTX-30xx is the best bet for now since laptop GPU yan hindi pang-desktop. Mas mababa yung specs at GPU benchmarks ng laptop versions compared sa desktop. It doesn't matter which CPU you use, dahil GPU applications ang gagamitin mo at the most. Alam ko merong laptop na RTX-3050 (pero 4gb vram lang?) na below 40K-50K with 16GB system ram. Pero kung desktop yung setup mo, ay bababa pa yang price at yung RTX-3050 is around 8gb vram na nasa safe side sa ganitong klaseng softwares. Balikan mo yung requirements to help you decide which to buy. Ito estimate ko sa laptops ngayon: 6GB vram GPU (50K-60K), 8GB GPU(60K-70K) yung pinakamura!

Ang ginagawa ko sa ngayon para magamit ko ng matino ay pinaghihiwalay ko yung features nila. Iba yung sa image generation, iba yung upscaling, face/body fixing, impainting/img2img, atbp. para di hirap yung hardware ko. AIO kasi yang SD kaya malakas yung kakaining memory kung ang daming gagawin. Marami dyan, plugins lang. Kaya lang, mauubos oras ko sa processing sa chop-chop SD project ko kung sakali he he.
Kaya huwag mong pilitin sa cpu-mode, magsisisi ka. Maraming options sa net kung gugustuhin mo lang kung sakaling kailangan mo talaga. PM mo ako kung gusto mong mag-test ng matagal habang wala ka pang hardware. Yon lang ang alam ko, hindi dyan sa image generation he he. Abala ako sa ibang AI projects para magamit ng anak ko sa pasukan. Pero, naghahanap pa rin ako ng pag-asa sa SD cpu-only development sa ngayon. Oras na may update at ok, i-post ko dito.

 

[alist1986]

UPDATES:
1. I tried Stable Diffusion Openvino using theYou do not have permission to view the full content of this post. Log in or register now.above. Works fine but you can only use some few settings compared sa original stable diffusion - just a prompt and upscaler option! Speed is slightly faster than Comfy and A1111. Buti pa sa free online service ka na lang gumawa kung positive prompt lang at upscaler yung gagamitin.
2. ComfyUI is the most stable sa lahat pero nahihirapan ako sa modular display kung saan ako mag-set he he. Pero nagagawa ko naman following these examples as guidance: You do not have permission to view the full content of this post. Log in or register now.
3. A1111 portable is slightly ok sa cpu-mode bukod sa user-friendly GUI vs. ComfyUI. - Best choice to try on my 12GB ram! Lahat ng new features supported nya pati extensions. Nasa user na lang yung pag-iinagat para di siya manghingalo sa settings na gagamitin. Di ko pa nasubukan sa training sa Controlnet.

 

[alist1986]

MariaGomez,​

Ang mainam is just buy an extra CPU ram at the moment, say 8 -16 GB for testing A1111 GUI to make it work for now. Ang next is try to open some sample images sa CIVITAI.com or replicate.com site and check the generation data for the sample pics. Study and collect all necessary models used including texture inversions or negative embeddings, etc. and try to reproduce the same image using the same seed. Marami pa kasing i-download if you are using SD offline. Nagsubok ako ngayon using A1111:

Wala pa akong upscaling dyan at kulang yung required embeddings sa negative prompts (ng_deepnegative_v1_75t,badhandv4) at majicMIX realistic f16 yung gamit ko, pero halos same di ba? Sa huli na lang yung upscaling for study!
(Yung left, yung original sa You do not have permission to view the full content of this post. Log in or register now. at yung right yung sample generation sa pc ko.)
Hindi masyadong matakaw sa ram siya sa txt2img generation. Makukuha sa step-by-step method using cpu-mode only. Mag-iintay ka lang ng 5 - 10 minutes.

 

[alist1986]

Yung nakasubok na ng SD sa GPU or CPU, pwede nyong gamitin yung You do not have permission to view the full content of this post. Log in or register now. as extension to create a talking avatar similar to You do not have permission to view the full content of this post. Log in or register now. for offline unlimited use.
Parehas lang yung process ng installing extensions sa GUI. Just follow the guide here:
You do not have permission to view the full content of this post. Log in or register now.

Download the external checkpoints including releases, set the path and continue with the rest of the guide.
Check nyo rin ito for additional guide to make it work for you: You do not have permission to view the full content of this post. Log in or register now.
For cpu mode users, ang isa pang idadagdag ninyo bukod doon sa binanggit ko noong una sa webui_user.bat launcher ay ito "--disable-safe-unpickle".
As usual, mabagal sa cpu-mode pero a working alternative.
Check nyo na lang dito sa github yung ibang paraan para mapagana ninyo siya w/ or w/out SD, online or offline:
You do not have permission to view the full content of this post. Log in or register now.
Yung naghahanap ng Google Colab option, try this guide:
You do not have permission to view the full content of this post. Log in or register now. (Di ko nasubukan ito)

 

[snailmanblue777]

pwede ba maglagay ng roop sa easy diffusion

 

[alist1986]

pwede ba maglagay ng roop sa easy diffusion

A1111 na lang natira sa'kin kaya di ko ma-verify sa ibang UI. Basta may option maglagay ng plugin o extension sa UI na gamit mo, kahit anong UI ay pwede. Kung ayaw, install ka ng standalone na roop.

Sa nabasa ko, may option to use extensions or plugins sa Easy-Diffusion as stated here:
You do not have permission to view the full content of this post. Log in or register now.
The file name must have the suffix .plugin.js. Di siya compatible sa standard format ng plugins for most Stabel Diffusion Uis like A1111 kaya mahirap maghanap ng roop plugin sa Easy diffusion. Install ka na lang ng SD A1111 or ComfyUI. Mas stable mga yan kaysa ED at mas mabilis kahit sa cpu-mode bukod sa maraming extensions na pwedeng ikabit - halos lahat.

Kung nahihirapan kang mag-install ng standard Stable Diffusion for some reason use You do not have permission to view the full content of this post. Log in or register now.. Ito yung upgraded version ng You do not have permission to view the full content of this post. Log in or register now. in "exe" format. Click and install lang.

Kung roop lang ang kailangan mo with no SD, install this: "You do not have permission to view the full content of this post. Log in or register now." Ito yung original roop na kinopya ng iba. Note. Maraming requirements ito. Another option is installing its clone "You do not have permission to view the full content of this post. Log in or register now.". Both require Visual studio 2022!

Hanap ka na lang dito:
1. You do not have permission to view the full content of this post. Log in or register now.
2. You do not have permission to view the full content of this post. Log in or register now.

 

[snailmanblue777]

A1111 na lang natira sa'kin kaya di ko ma-verify sa ibang UI. Basta may option maglagay ng plugin o extension sa UI na gamit mo, kahit anong UI ay pwede. Kung ayaw, install ka ng standalone na roop.

Sa nabasa ko, may option to use extensions or plugins sa Easy-Diffusion as stated here:
You do not have permission to view the full content of this post. Log in or register now.
The file name must have the suffix .plugin.js. Di siya compatible sa standard format ng plugins for most Stabel Diffusion Uis like A1111 kaya mahirap maghanap ng roop plugin sa Easy diffusion. Install ka na lang ng SD A1111 or ComfyUI. Mas stable mga yan kaysa ED at mas mabilis kahit sa cpu-mode bukod sa maraming extensions na pwedeng ikabit - halos lahat.

Kung nahihirapan kang mag-install ng standard Stable Diffusion for some reason use You do not have permission to view the full content of this post. Log in or register now.. Ito yung upgraded version ng You do not have permission to view the full content of this post. Log in or register now. in "exe" format. Click and install lang.

Kung roop lang ang kailangan mo with no SD, install this: "You do not have permission to view the full content of this post. Log in or register now." Ito yung original roop na kinopya ng iba. Note. Maraming requirements ito. Another option is installing its clone "You do not have permission to view the full content of this post. Log in or register now.". Both require Visual studio 2022!

Hanap ka na lang dito:
1. You do not have permission to view the full content of this post. Log in or register now.
2. You do not have permission to view the full content of this post. Log in or register now.

Nainstall ko na roop sa a111 ko kaso walang lumalabas na ui sa SD ko kahit naka visual studio 2022 na ako atsaka naka python na din

 

[alist1986]

Nainstall ko na roop sa a111 ko kaso walang lumalabas na ui sa SD ko kahit naka visual studio 2022 na ako atsaka naka python na din

Kung sinunod mo yung guide sa "You do not have permission to view the full content of this post. Log in or register now." at walang errors, dapat makita mo yan sa UI na pinakita dyan. Yung binigay ko sa last post is standalone! Yung extension w/ SD, ayos lang gamitin sa cpu mode dahil for images yan pero yung standalone medyo mahirap dahil may video features yan kaya tinanggal ko na at malimit mag-hang sa booting pa lang kung naka-open browser ko he he.
I-reset mo muna yung A1111 para sigurado sa bawat install mo ng mga extensions. Check mo itong guide na may video, baka makatulong:
You do not have permission to view the full content of this post. Log in or register now.
You do not have permission to view the full content of this post. Log in or register now.
Alam mo na rin siguro ang difference ng pag-install using GPU and using CPU.
Check mo sa issues section ng extension sa github yung iba pang problema:
You do not have permission to view the full content of this post. Log in or register now.
Every pc has its own unique issue to fix in some cases. Dyan ako natuto.