CreatorsOk
Furkan Gözükara

Furkan Gözükara

patreon


Furkan Gözükara posts

SOTA Image Captioning Scripts For Stable Diffusion: CogVLM, LLaVA, BLIP-2, Clip-Interrogator (115 Clip Vision Models + 5 Caption Models)

Patreon exclusive posts index to find our scripts easily, Patreon scripts updates history to see which updates arrived to which scripts and amazing 2024-10-02 11:00:00 +0000 UTC View Post

MimicPC Tutorial - Easy Cloud AI Running Platform - Online AI Generator

What is MimicPC

  • MimicPC is a cloud service that lets you use AI applications seamlessly in your browser while everything runs on remote cloud servers, providing permanent storage as well

  • Full tutorial link : will be added hopefully once published

  • MimicPC Official Site : https://home.mimicpc.com?fpr=6qvql

  • MimicPC Official Discord server where y...

    View Post

Detailed Comparison of JoyCaption Alpha One vs JoyCaption Pre-Alpha - 10 Different Style Amazing Images - I think JoyCaption Alpha One is the very best image captioning model at the moment for model training - Works very fast and requires as low as 8.5 GB VRAM

Where To Download And Install

Multi-GPU FLUX Full Fine Tuning Experiments and Requirements on RunPod and Conclusions - Used 2x A100 - 80 GB GPUs

Patreon exclusive posts index to find our scripts easily, Patreon scripts updates history to see which updates arrived to which scripts and amazing 2024-09-21 15:31:55 +0000 UTC View Post

Single Block / Layer FLUX LoRA Training Research Results and LoRA Network Alpha Change Impact With LoRA Network Rank Dimension

Patreon exclusive posts index to find our scripts easily, Patreon scripts updates history to see which updates arrived to which scripts and amazing 2024-09-20 00:49:09 +0000 UTC View Post

Full Fine Tuning / DreamBooth of FLUX yields way better results than LoRA training as expected, overfitting and bleeding reduced a lot, check oldest comment for more information, images LoRA vs Fine Tuned full checkpoint

Full Fine Tuning / DreamBooth of FLUX yields way better results than LoRA training as expected, overfitting and bleeding reduced a lot, check oldest comment for more information, images LoRA vs Fine Tuned full checkpoint

Configs and Full Experiments

Trained Myself With 256 Images on FLUX - Results Mind Blowing

Detailed Full Workflow

Training a FLUX LoRA with 256 Images Experiments - Full Workflow and Conclusions

Patreon exclusive posts index to find our scripts easily, Patreon scripts updates history to see which updates arrived to which scripts and amazing 2024-09-13 15:33:00 +0000 UTC View Post

Massive 256 Images Training Started on 8x GPU

Started training my 256 images on FLUX with 8x GPU - Dataset is not ready yet, not very good sharpness and lightning but so many people asking expressions so i am taking a break from research :) - Going up to 200 epochs I also wonder results

View Post

SOTA Subject Cropper and Face Focused Image Resizer Scripts Do Better Training

16 September 2024

NOW WE HAVE A MUCH BETTER NEW APP PLEASE USE IT

https://www.patreon.com/posts/112126955

NEW APP : https://www.patreon.com/posts/112126955

THIS POST WONT BE MAINTAINED ANYMORE

USE NEW APP ...

View Post

OneTrainer FLUX LoRA Training Full Tutorial For Local Windows and Cloud Massed Compute

Patreon exclusive posts index to find our scripts easily, Patreon scripts updates history to see which updates arrived to which scripts and amazing 2024-09-10 16:49:36 +0000 UTC View Post

Ultimate FLUX LoRA Training Tutorial: Windows and Cloud Deployment

I have done total 104 different LoRA trainings and compared each one of them to find the very best hyper parameters and the workflow for FLUX LoRA training by using Kohya GUI training script.

You can see all the done experiments’ checkpoint names and their repo links in following public post: https://www.patreon.com/posts/110838414

After completing all these FLUX LoRA trainings by using the most VRAM...

View Post

All The LoRA FLUX Training Experiments I Have Done So Far

I have been using a machine on RunPod that has 8x RTX A6000 over a week now. I have done all the following experiments with each one being 3000 steps.

Latest configs shared on : https://www.patreon.com/posts/kohya-flux-lora-110879657

All findings, grids, and best configs and 1-click installers for Windows, RunPod and Massed Compute shared here : View Post

SECourses 3D Render for FLUX LoRA Model Published on CivitAI - Style Consistency Achieved - Full Workflow Shared on Hugging Face With Results of Experiments - Last Image Is Used Dataset


CivitAI Link : https://civitai.com/models/731347

Hugging Face Link : https://huggingface.co/MonsterMMORPG/3D-Cartoon-Style-FLUX

SECourses 3D Render for FLUX LoRA Model Published on CivitAI - Style Consistency Achieved - Full Workflow Shared on Huggi...

View Post

Rope NEXT Update Has Arrived - both Rope Live and Rope Alucard Versions Are Updated

  • Rope NEXT Update Has Arrived - both Rope Live and Rope Alucard Versions Are Updated

  • Both works with CUDA 11.8 and also CUDA 12.4

  • Both works on Massed Compute perfectly

  • You can download installer zip files here with instructions : https://www.patreon.com/posts/105123768

  • Now works faster and better than before

  • 1-Click...

    View Post

1-Click Install FaceFusion and FaceFusion NEXT on Windows, RunPod and Massed Compute

The installer zip files are provided in this post : https://www.patreon.com/posts/103765029

It took huge time to make FaceFusion NEXT work with onnx

Requirements

Raw FLUX LoRA Output vs SwarmUI Quick 2x Upscale Button vs Our SUPIR APP Upscale (Face Upscale ON)

Raw FLUX LoRA Output vs SwarmUI Quick 2x Upscale Button vs Our SUPIR APP Upscale (Face Upscale ON) : https://imgsli.com/MjkzODAz/2/1

Our SUPIR APP : https://youtu.be/OYxVEvDf284

I am batch upscaling some images to add intro part of the Massed Compute and RunPod FLUX LoRA cloud training tutorial. Hopefully almost ready.

View Post

Training A Public FLUX Style LoRA - All Process Will Be Shared

I started training a public LoRA style (2 seperate training each on 4x A6000).

Experimenting captions vs non-captions. So we will see which yields best results for style training on FLUX.

Generated captions with multi-GPU batch Joycaption app.

I am showing 5 examples of what Joycaption generates on FLUX dev. Left images are the original style images from the dataset.

I used my multi-GPU Joycaption APP (used 8x A6000 for ultra fast captioning) : 2024-09-02 23:48:09 +0000 UTC View Post

Kohya FLUX LoRA and Fine Tuning Training Full Tutorial For Local Windows and Cloud RunPod and Massed Compute for Research & Development

I will update this page as I progress. I am just starting so I will add the info. Nothing ready yet

FOLLOW THIS THREAD NOW FOR TUTORIAL : https://www.patreon.com/posts/110879657

Who has fbgemm.dll error on Windows 11 the solutions posted below topic

First fully multi-GPU supporting and very advanced batch image captioner APP with Gradio interface published

Multi-GPU batch caption with JoyCaption. JoyCaption uses Meta-Llama-3.1–8B and google/siglip-so400m-patch14–384 and a fine tuned image captioning neural network.

Link : https://www.patreon.com/posts/110613301

Link for batch caption editor : https://www.patreon.com/posts/108992085

Coding multi-gpu in Python and Tor...

View Post

JoyCaption is amazing to caption training data. Advanced Gradio APP with batch processing and 1-click install


Here a Hugging Face space that you can test it yourself : https://huggingface.co/spaces/fancyfeast/joy-caption-pre-alpha - still working

I have been requested to make a Gradio app for this so i made an advanced app and 1-click installers

It uses a clip siglip-so400m-patch14-384 and Meta-Llama-3.1-8B-Instruct as model and a fine tuned checkpoint for better cap...

View Post

ResShift 1-Click Windows, RunPod, Massed Compute, Kaggle Installers with Amazing Gradio APP and Batch Image Processing. ResShift is Efficient Diffusion Model for Image Super-resolution by Residual Shifting (NeurIPS 2023, Spotlight).

ResShift: Efficient Diffusion Model for Image Super-resolution by Residual Shifting (NeurIPS 2023, Spotlight)

Official Repo : https://github.com/zsyOAOA/ResShift

I have developed a very advanced Gradio APP.

Developed APP Scripts and Installers : https://www.patreon.com/post...

View Post

ResShift 1-Click Windows, RunPod, Massed Compute, Kaggle Installers with Amazing Gradio APP and Batch Image Processing

Patreon exclusive posts index to find our scripts easily, Patreon scripts updates history to see which updates arrived to which scripts and amazing 2024-08-18 21:57:40 +0000 UTC View Post

Forge Web UI Latest Version RunPod Auto Installer and FLUX Auto Model Downloader for Windows, RunPod and Massed Compute

Scripts are shared here : https://www.patreon.com/posts/forge-web-ui-and-110323512

Uses official Pytorch 11.8 template of RunPod

The command line arguments are also automatically optimized for 24 GB and above VRAM GPUs for FP8 DEV model

No gpu offloading or reloading is made

Below speeds for single RTX 4090 GPU

1.6 it / second for 1024x1024 pixel resolution for FP8 DE...

View Post

Forge Web UI Latest Version RunPod Auto Installer and FLUX Auto Model Downloader for Windows, RunPod and Massed Compute

this post is not maintained anymore here newer version : https://www.patreon.com/posts/118442039

this post is not maintained anymore here newer version : https://www.patreon.com/posts/118442039

this post is not maintained anymore here newer version : 2024-08-18 12:28:23 +0000 UTC View Post

The Very Best Workflow For SDXL DreamBooth / Full Fine Tuning - Results Of 100+ Full Trainings

Patreon exclusive posts index to find our scripts easily, Patreon scripts updates history to see which updates arrived to which scripts and amazing 2024-08-16 22:00:00 +0000 UTC View Post

Comprehensive Different Version and Precision FLUX Models Speed and VRAM Usage Comparison

I just updated the automatic FLUX models downloader scripts with newest models and features. Therefore I decided to test all models comprehensively with respected to their peak VRAM usage and also their image generation speed.

Automatic downloader scripts : https://www.patreon.com/posts/109289967

 2024-08-15 03:25:49 +0000 UTC View Post

Full Tutorial for More Advanced SUPIR Online from Original Authors by SupPixel AI

I have been recently contacted by the original SUPIR developers that they published SUPIR online with better features and they are generated funds from this service to develop SUPIR v2 and SUPIR video.

As you know, SUPIR is the best image upscaler available right now both commercially and open source. It is next level compared to all other upscalers. So I decided to prepare this tutorial video and post to help SUPIR developers.

Tutorial video link : will be added when published View Post

FLUX Models 1-Click Auto Downloaders for SwarmUI for Windows, RunPod and Massed Compute

Patreon exclusive posts index to find our scripts easily, Patreon scripts updates history to see which updates arrived to which scripts and amazing 2024-08-14 21:11:00 +0000 UTC View Post

AuraSR GigaGAN 4x Upscaler Local (Windows), Cloud (RunPod, Massed Compute, Kaggle) 1-Click Installers and Batch Processing App

AuraSR is a 600M parameter upsampler model derived from the GigaGAN paper. It works super fast and uses a very limited VRAM below 5 GB. It is deterministic upscaler. It works perfect in some images but fails in some images so it is worth to give it a shot.

GitHub official repo : https://github.com/fal-ai/aura-sr

I have developed 1-click installers and a batch upscaler App.

You can download installers a...

View Post