SUPIR Full Tutorial + 1 Click 12GB VRAM Windows & RunPod / Linux Installer + Batch Upscale + Comparison With Magnific - SUPIR Starts A New Era
•
1
Furkan Gözükara
MonsterMMORPG
AI & ML interests
Check out my youtube page SECourses for Stable Diffusion tutorials. They will help you tremendously in every topic
Articles
Organizations
MonsterMMORPG's activity
posted
an
update
8 days ago
Post
2736
Complete Guide to SUPIR Enhancing and Upscaling Images Like in Sci-Fi Movies on Your PC : https://youtu.be/OYxVEvDf284
In this video, I explain how to 1 click install and use the most advanced image upscaler / enhancer in the world that is both commercially and open source available. The upscaler that I am going to introduce you is open source #SUPIR and the model is free to use. SUPIR upscaler is many times better than both paid Topaz AI and Magnific AI and you can use this upscaler on your computer for free forever. The difference of SUPIR vs #Topaz and #Magnific is like ages. So in this tutorial you are going to learn everything about how to install, update and use SUPIR upscaler on your personal computer. The video shows Windows but it works perfectly fine on Linux as well.
Scripts Download Link ⤵️
https://www.patreon.com/posts/99176057
Samplers and Text CFG (Text Guidance Scale) Comparison Link ⤵️
https://imgsli.com/MjU2ODQz/2/1
How to install accurate Python, Git and FFmpeg on Windows Tutorial ⤵️
https://youtu.be/-NjNy7afOQ0
Full DreamBooth / Fine-tuning Tutorial ⤵️
https://youtu.be/0t5l6CP9eBg
Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild : https://arxiv.org/abs/2401.13627
Authors introduce SUPIR (Scaling-UP Image Restoration), a groundbreaking image restoration method that harnesses generative prior and the power of model scaling up. Leveraging multi-modal techniques and advanced generative prior, SUPIR marks a significant advance in intelligent and realistic image restoration. As a pivotal catalyst within SUPIR, model scaling dramatically enhances its capabilities and demonstrates new potential for image restoration. Authors collect a dataset comprising 20 million high-resolution, high-quality images for model training, each enriched with descriptive text annotations. SUPIR provides the capability to restore images guided by textual prompts, broadening its application scope and potential
In this video, I explain how to 1 click install and use the most advanced image upscaler / enhancer in the world that is both commercially and open source available. The upscaler that I am going to introduce you is open source #SUPIR and the model is free to use. SUPIR upscaler is many times better than both paid Topaz AI and Magnific AI and you can use this upscaler on your computer for free forever. The difference of SUPIR vs #Topaz and #Magnific is like ages. So in this tutorial you are going to learn everything about how to install, update and use SUPIR upscaler on your personal computer. The video shows Windows but it works perfectly fine on Linux as well.
Scripts Download Link ⤵️
https://www.patreon.com/posts/99176057
Samplers and Text CFG (Text Guidance Scale) Comparison Link ⤵️
https://imgsli.com/MjU2ODQz/2/1
How to install accurate Python, Git and FFmpeg on Windows Tutorial ⤵️
https://youtu.be/-NjNy7afOQ0
Full DreamBooth / Fine-tuning Tutorial ⤵️
https://youtu.be/0t5l6CP9eBg
Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild : https://arxiv.org/abs/2401.13627
Authors introduce SUPIR (Scaling-UP Image Restoration), a groundbreaking image restoration method that harnesses generative prior and the power of model scaling up. Leveraging multi-modal techniques and advanced generative prior, SUPIR marks a significant advance in intelligent and realistic image restoration. As a pivotal catalyst within SUPIR, model scaling dramatically enhances its capabilities and demonstrates new potential for image restoration. Authors collect a dataset comprising 20 million high-resolution, high-quality images for model training, each enriched with descriptive text annotations. SUPIR provides the capability to restore images guided by textual prompts, broadening its application scope and potential
posted
an
update
18 days ago
Post
3605
Watch the full tutorial here : https://youtu.be/0t5l6CP9eBg
The tutorial is over 2 hours literally with manually fixed captions and perfect video chapters.
Most Awaited Full Fine Tuning (with DreamBooth effect) Tutorial Generated Images - Full Workflow Shared In The Comments - NO Paywall This Time - Explained OneTrainer - Cumulative Experience of 16 Months Stable Diffusion
In this tutorial, I am going to show you how to install OneTrainer from scratch on your computer and do a Stable Diffusion SDXL (Full Fine-Tuning 10.3 GB VRAM) and SD 1.5 (Full Fine-Tuning 7GB VRAM) based models training on your computer and also do the same training on a very cheap cloud machine from MassedCompute if you don't have such computer.
Tutorial Readme File ⤵️
https://github.com/FurkanGozukara/Stable-Diffusion/blob/main/Tutorials/OneTrainer-Master-SD-1_5-SDXL-Windows-Cloud-Tutorial.md
Register Massed Compute From Below Link (could be necessary to use our Special Coupon for A6000 GPU for 31 cents per hour) ⤵️
https://bit.ly/Furkan-Gözükara
Coupon Code for A6000 GPU is : SECourses
0:00 Introduction to Zero-to-Hero Stable Diffusion (SD) Fine-Tuning with OneTrainer (OT) tutorial
3:54 Intro to instructions GitHub readme
4:32 How to register Massed Compute (MC) and start virtual machine (VM)
5:48 Which template to choose on MC
6:36 How to apply MC coupon
8:41 How to install OT on your computer to train
9:15 How to verify your Python, Git, FFmpeg and Git installation
12:00 How to install ThinLinc and start using your MC VM
12:26 How to setup folder synchronization and file sharing between your computer and MC VM
13:56 End existing session in ThinClient
14:06 How to turn off MC VM
14:24 How to connect and start using VM
14:41 When use end existing session
16:38 How to download very best OT preset training configuration for SD 1.5 & SDXL models
18:00 How to load configuration preset
18:38 Full explanation of OT configuration and best hyper parameters for SDXL
.
.
.
The tutorial is over 2 hours literally with manually fixed captions and perfect video chapters.
Most Awaited Full Fine Tuning (with DreamBooth effect) Tutorial Generated Images - Full Workflow Shared In The Comments - NO Paywall This Time - Explained OneTrainer - Cumulative Experience of 16 Months Stable Diffusion
In this tutorial, I am going to show you how to install OneTrainer from scratch on your computer and do a Stable Diffusion SDXL (Full Fine-Tuning 10.3 GB VRAM) and SD 1.5 (Full Fine-Tuning 7GB VRAM) based models training on your computer and also do the same training on a very cheap cloud machine from MassedCompute if you don't have such computer.
Tutorial Readme File ⤵️
https://github.com/FurkanGozukara/Stable-Diffusion/blob/main/Tutorials/OneTrainer-Master-SD-1_5-SDXL-Windows-Cloud-Tutorial.md
Register Massed Compute From Below Link (could be necessary to use our Special Coupon for A6000 GPU for 31 cents per hour) ⤵️
https://bit.ly/Furkan-Gözükara
Coupon Code for A6000 GPU is : SECourses
0:00 Introduction to Zero-to-Hero Stable Diffusion (SD) Fine-Tuning with OneTrainer (OT) tutorial
3:54 Intro to instructions GitHub readme
4:32 How to register Massed Compute (MC) and start virtual machine (VM)
5:48 Which template to choose on MC
6:36 How to apply MC coupon
8:41 How to install OT on your computer to train
9:15 How to verify your Python, Git, FFmpeg and Git installation
12:00 How to install ThinLinc and start using your MC VM
12:26 How to setup folder synchronization and file sharing between your computer and MC VM
13:56 End existing session in ThinClient
14:06 How to turn off MC VM
14:24 How to connect and start using VM
14:41 When use end existing session
16:38 How to download very best OT preset training configuration for SD 1.5 & SDXL models
18:00 How to load configuration preset
18:38 Full explanation of OT configuration and best hyper parameters for SDXL
.
.
.
replied to
their
post
30 days ago
thanks a lot
posted
an
update
about 1 month ago
Post
1959
Compared Effect Of Image Captioning For SDXL Fine-tuning / DreamBooth Training for a Single Person, 10.3 GB VRAM via OneTrainer
Sadly the post character count is limited so please read full info on Medium here
https://medium.com/@furkangozukara/compared-effect-of-image-captioning-for-sdxl-fine-tuning-dreambooth-training-for-a-single-person-961087e42334
Sadly the post character count is limited so please read full info on Medium here
https://medium.com/@furkangozukara/compared-effect-of-image-captioning-for-sdxl-fine-tuning-dreambooth-training-for-a-single-person-961087e42334
replied to
their
post
about 1 month ago
I did over 100 trainings empirically to find best hyper parameters. And training U-NET + Text Encoder 1 yields better results that only U-NET @researcher171473
posted
an
update
about 1 month ago
Post
1773
Now You Can Full Fine Tune / DreamBooth Stable Diffusion XL (SDXL) with only 10.3 GB VRAM via OneTrainer — Both U-NET and Text Encoder 1 is trained — Compared 14 GB config vs slower 10.3 GB Config
Full config and instructions are shared here : https://www.patreon.com/posts/96028218
Used SG161222/RealVisXL_V4.0 as a base model and OneTrainer to train on Windows 10 : https://github.com/Nerogar/OneTrainer
The posted example x/y/z checkpoint comparison images are not cherry picked. So I can get perfect images with multiple tries.
Trained 150 epochs, 15 images and used my ground truth 5200 regularization images : https://www.patreon.com/posts/massive-4k-woman-87700469
In each epoch only 15 of regularization images used to make DreamBooth training affect
As a caption only “ohwx man” is used, for regularization images just “man”
You can download configs and full instructions here : https://www.patreon.com/posts/96028218
Hopefully full public tutorial coming within 2 weeks. I will show all configuration as well
The tutorial will be on our channel : https://www.youtube.com/SECourses
Training speeds are as below thus durations:
RTX 3060 — slow preset : 3.72 second / it thus 15 train images 150 epoch 2 (reg images concept) : 4500 steps = 4500 3.72 / 3600 = 4.6 hours
RTX 3090 TI — slow preset : 1.58 second / it thus : 4500 * 1.58 / 3600 = 2 hours
RTX 3090 TI — fast preset : 1.45 second / it thus : 4500 * 1.45 / 3600 = 1.8 hours
A quick tutorial for how to use concepts in OneTrainer : https://youtu.be/yPOadldf6bI
Full config and instructions are shared here : https://www.patreon.com/posts/96028218
Used SG161222/RealVisXL_V4.0 as a base model and OneTrainer to train on Windows 10 : https://github.com/Nerogar/OneTrainer
The posted example x/y/z checkpoint comparison images are not cherry picked. So I can get perfect images with multiple tries.
Trained 150 epochs, 15 images and used my ground truth 5200 regularization images : https://www.patreon.com/posts/massive-4k-woman-87700469
In each epoch only 15 of regularization images used to make DreamBooth training affect
As a caption only “ohwx man” is used, for regularization images just “man”
You can download configs and full instructions here : https://www.patreon.com/posts/96028218
Hopefully full public tutorial coming within 2 weeks. I will show all configuration as well
The tutorial will be on our channel : https://www.youtube.com/SECourses
Training speeds are as below thus durations:
RTX 3060 — slow preset : 3.72 second / it thus 15 train images 150 epoch 2 (reg images concept) : 4500 steps = 4500 3.72 / 3600 = 4.6 hours
RTX 3090 TI — slow preset : 1.58 second / it thus : 4500 * 1.58 / 3600 = 2 hours
RTX 3090 TI — fast preset : 1.45 second / it thus : 4500 * 1.45 / 3600 = 1.8 hours
A quick tutorial for how to use concepts in OneTrainer : https://youtu.be/yPOadldf6bI
-
- 5 replies
replied to
their
post
2 months ago
100% this is next level. thanks for comment @ajibawa-2023
replied to
their
post
2 months ago
@ameerazam08 100%. I am talking with original developers for CPU Offloading too if they hopefully add.
posted
an
update
2 months ago
Post
I have dedicated several days, working over 12 hours each day, on SUPIR (Scaling-UP Image Restoration), a cutting-edge image enhancement and upscaling model introduced in the paper Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild.
This model is simply mind-blowing. At the bottom of this post, you will see side-by-side comparisons of SUPIR versus the extremely expensive online service, Magnific AI. Magnific is known to be the best among the community. However, SUPIR is by far superior. SUPIR also significantly outperforms Topaz AI upscale. SUPIR manages to remain faithful to the original image almost 100% while adding details and achieving super upscaling with the best realism.
You can read the full blog post here : https://huggingface.co/blog/MonsterMMORPG/supir-sota-image-upscale-better-than-magnific-ai
This model is simply mind-blowing. At the bottom of this post, you will see side-by-side comparisons of SUPIR versus the extremely expensive online service, Magnific AI. Magnific is known to be the best among the community. However, SUPIR is by far superior. SUPIR also significantly outperforms Topaz AI upscale. SUPIR manages to remain faithful to the original image almost 100% while adding details and achieving super upscaling with the best realism.
You can read the full blog post here : https://huggingface.co/blog/MonsterMMORPG/supir-sota-image-upscale-better-than-magnific-ai
-
- 4 replies
posted
an
update
2 months ago
Post
Compared Stable Diffusion 3 with Dall-E3 and results are mind blowing.
SD 3 can follow prompts many times better than SD 1.5 or SDXL. It is even better than Dall-E3 in following text / spelling prompts.
The realism of the SD 3 can't be even compared with Dall-E3, since every Dall-E3 output is like a digital render.
Can't wait to get approved of Stability AI early preview program to do more intensive testing.
Some people says be skeptical about cherry picking. I agree but I hope that these Stability AI released images are not that heavy cherry picking.
You can see SD3 vs Dall-E3 comparison here : https://youtu.be/DJxodszsERo
SD 3 can follow prompts many times better than SD 1.5 or SDXL. It is even better than Dall-E3 in following text / spelling prompts.
The realism of the SD 3 can't be even compared with Dall-E3, since every Dall-E3 output is like a digital render.
Can't wait to get approved of Stability AI early preview program to do more intensive testing.
Some people says be skeptical about cherry picking. I agree but I hope that these Stability AI released images are not that heavy cherry picking.
You can see SD3 vs Dall-E3 comparison here : https://youtu.be/DJxodszsERo
posted
an
update
3 months ago
Post
Today my RunPod pod was broken and I didn't notice until I fully did setup it. So I have written the following tutorial for how to deploy a Pod and also verify it is not broken.
You can read on:
Patreon (public) : https://www.patreon.com/posts/how-to-deploy-on-97919576
Medium (public) : https://medium.com/@furkangozukara/how-to-deploy-a-pod-on-runpod-and-verify-it-is-working-20e47031c0b5
CivitAI (public) : https://civitai.com/articles/3994/how-to-deploy-a-pod-on-runpod-and-verify-it-is-working
LinkedIn (public) : https://www.linkedin.com/pulse/how-deploy-pod-runpod-verify-working-furkan-g%2525C3%2525B6z%2525C3%2525BCkara-lgplf%3FtrackingId=EuNOjpKCSQ%252BVfpiQV3D6KQ%253D%253D/?trackingId=EuNOjpKCSQ%2BVfpiQV3D6KQ%3D%3D
Dev . to (public) : https://dev.to/furkangozukara/how-to-deploy-a-pod-on-runpod-and-verify-it-is-working-3pop
You can read on:
Patreon (public) : https://www.patreon.com/posts/how-to-deploy-on-97919576
Medium (public) : https://medium.com/@furkangozukara/how-to-deploy-a-pod-on-runpod-and-verify-it-is-working-20e47031c0b5
CivitAI (public) : https://civitai.com/articles/3994/how-to-deploy-a-pod-on-runpod-and-verify-it-is-working
LinkedIn (public) : https://www.linkedin.com/pulse/how-deploy-pod-runpod-verify-working-furkan-g%2525C3%2525B6z%2525C3%2525BCkara-lgplf%3FtrackingId=EuNOjpKCSQ%252BVfpiQV3D6KQ%253D%253D/?trackingId=EuNOjpKCSQ%2BVfpiQV3D6KQ%3D%3D
Dev . to (public) : https://dev.to/furkangozukara/how-to-deploy-a-pod-on-runpod-and-verify-it-is-working-3pop