LoRA and Fine-Tuning Training: Kohya, Musubi Tuner, OneTrainer, AI Toolkit
Train your own LoRAs and full checkpoints (FLUX, Qwen Image, Wan, SDXL, SD 1.5): 1-click trainer installers, the best configs, full tutorials and our training research.

Musubi Tuner App: 1-Click LoRA & Fine-Tune Training With 435 Presets
SECourses Musubi Tuner trains LoRAs and full fine-tunes of Qwen Image, Wan 2.2, FLUX 2, Z Image and LTX 2.3 on Windows or cloud GPUs with 435 ready presets.

Ostris AI-Toolkit 1-Click Installers: LoRA Training on Windows & Cloud
Ostris AI-Toolkit installs automatically on Windows, RunPod, SimplePod and Massed Compute and is now patched to train on custom single-file base models.

Qwen Image 2512 Training Comparison: The Best Model to Train
Qwen Image 2512 is the new king for training: the video compares trained Base, 2512, Edit 2509 and Edit 2511 models and shows a big leap in realism.

Qwen Image 2512 First Training: 120 vs 180 vs 240 Epochs Comparison
Qwen Image 2512 first training results: the model behaves like FLUX Dev, and more training makes it more realistic with less noise on a 28-image dataset.

Wan 2.2 LoRA Training Tutorial: Text to Image & Video, Windows & Cloud
Wan 2.2 complete LoRA training tutorial with Musubi Tuner, based on over 64 trainings; the ready configs train locally on GPUs with as little as 6 GB VRAM.

Z-Image Turbo Tutorial: LoRA Training, ControlNet Union & Presets
Z-Image Turbo step by step: generate with ready SwarmUI presets, add Fun ControlNet Union and train LoRAs with Ostris AI Toolkit on GPUs from 6 GB.

Kohya FLUX LoRA Training: Full Tutorial & Configs for Windows & Cloud
Kohya SS GUI FLUX LoRA training step by step, with the best optimized configs for 4 GB to 48 GB GPUs and a 68 minute Windows tutorial with 74 chapters.

Kohya FLUX Fine Tuning / DreamBooth Training Tutorial: Windows & Cloud
Kohya FLUX full fine tuning / DreamBooth trains full checkpoints with configs for GPUs with as low as 8 GB VRAM, all using Torch Compile on Windows and Linux.

FLUX 2 vs FLUX SRPO & New Kohya SS GUI Premium FLUX Training App
FLUX 2 Pro compared with a trained FLUX SRPO model, plus the new Kohya trainer V35 that trains FLUX on GPUs with as little as 6 GB VRAM.

Qwen Image vs FLUX SRPO Training Comparison: 20 Images, Same Dataset
Qwen Image base model vs FLUX SRPO training on the same 28-image dataset: 20 side-by-side results show Qwen far ahead at complex prompts and emotions.

Qwen Image Training Tutorial: LoRA & Fine Tuning, Base & Edit Models
Qwen Image LoRA and full fine-tuning tutorial with SECourses Musubi Tuner, built on 21 days of research; trains on Windows GPUs with as little as 6 GB VRAM.

Claude AI Edits Word & Excel Files, Plus Wan Training Arrives
Claude AI now edits Word documents and Excel spreadsheets; the video also shows Wan and Qwen Image training in Musubi Tuner and IndexTTS2 voice cloning.

Qwen Image LoRA Training: Stage 1 Results and Pre-Made Configs
Qwen Image LoRA results from more than 35 Stage 1 trainings with SECourses Musubi Tuner; the published pre-made configs train on GPUs with as low as 6 GB.

Decoding the Timestep Shift in Diffusion Training: Qwen Image & FLUX
Timestep sampling shift explained for diffusion models: Qwen Image uses an exponential shift and FLUX a linear one, and 2.205 matches Qwen at 1664x928.

Kohya FLUX DreamBooth Results: 20 Profile Images Generated Locally
Kohya FLUX DreamBooth showcase: 20 profile images generated recently in SwarmUI to change my profile photo, each 2x latent upscaled to 4 megapixels.

Kohya vs OneTrainer for FLUX Dev Fine-Tuning / DreamBooth: Comparison
Kohya and OneTrainer compared for FLUX Dev fine-tuning: OneTrainer with EMA and a copied Kohya config gave weaker results than the best Kohya config.

Kohya GUI: Convert FLUX, SD 3.5 & SDXL LoRAs to FP8 to Save Space
Converting a trained LoRA to FP8 with Kohya GUI shrinks a 2.5 GB FP32 file to 634 MB with almost the best quality, as the FP8 vs FP16 vs FP32 grid shows.

FLUX Fine-Tuning / DreamBooth Images: 2x Latent Upscale, 2048x2048
FLUX fine-tuning / DreamBooth images made with a simple prompt and 2x latent upscale; each is 2048x2048 and training works on as low as 6 GB GPUs.

Kohya FLUX LoRA on 4 GB and DreamBooth / Fine-Tuning on 6 GB GPUs
Kohya now trains FLUX LoRAs on 4 GB GPUs and full FLUX DreamBooth / fine-tuning on 6 GB, thanks to faster block swapping, FP8 and training only certain layers.

FLUX Fine-Tuning vs LoRA: Example Dataset, Trained Models & Configs
FLUX fine-tuning / DreamBooth vs LoRA experiment shared in full: a 28-image dataset, Kohya configs, grids and every trained checkpoint free on Hugging Face.

Perfect Quality Training Image Dataset for FLUX, SD 3.5, SDXL & SD 1.5
Example model training dataset of 28 maximum-quality 1024x1024 images, prepared to learn and test training, sized for the FLUX batch size 7 config.

FLUX De-Distilled Fine-Tuning & LoRA Tests: Bleeding and CFG Results
FLUX de-distilled models train best at guidance scale 3.5 instead of 1, yet these DreamBooth and LoRA experiments show they still do not fix bleeding.

FLUX DreamBooth Checkpoints: Convert 24 GB FP16 to 12 GB FP8 Tutorial
FLUX fine tuning / DreamBooth checkpoints take 23.8 GB in FP16; Kohya GUI converts them to 11.9 GB FP8 with no visible quality loss in the grids.

FLUX LoRA vs Fine Tuning / DreamBooth Comparison: Batch Size 1 vs 7
FLUX LoRA and Fine Tuning / DreamBooth compared with 15 and 256 training images at batch size 1 and 7, for both realism and stylization.

Kohya GUI: Full FLUX Dev DreamBooth on 6 GB GPUs Without Quality Loss
Kohya GUI can now fully fine-tune FLUX Dev on 6 GB GPUs with the same quality as 48 GB GPUs, and fine-tuning gives better results than any LoRA training.

Multi-GPU FLUX Full Fine-Tuning Test on RunPod: 2x A100 80 GB Results
Multi-GPU FLUX full fine-tuning / DreamBooth on 2x A100 80 GB GPUs gave only a 26% speed gain over one GPU, so 2x GPU training is not worth it yet.

FLUX LoRA Single Block Training Research: Network Alpha & Rank Tests
FLUX LoRA single block / layer training tested in Kohya GUI: one block needed over 10 times the usual learning rate and still gave sub-par results.

FLUX Full Fine Tuning vs LoRA Comparison: Less Overfitting & Bleeding
FLUX full fine tuning / DreamBooth gives way better results than LoRA training, as expected, with much less overfitting and bleeding.

FLUX LoRA Trained on 256 Images of Myself: Mind-Blowing Results
FLUX LoRA trained with Kohya GUI on 256 phone photos of myself, including the Clip-L and T5-XXL text encoders; it learned body shape and fine facial details.

FLUX LoRA Training With 256 Images: Full Workflow & Conclusions
FLUX LoRA training experiments with 256 images on 8x A6000 and 1x L40S GPUs: 80 epochs worked best, and doubling the learning rate for 8 GPUs overtrained.

Kohya FLUX LoRA Training Experiments: All 104 Trainings Listed
FLUX LoRA training experiments run with Kohya on an 8x RTX A6000 RunPod machine, 3000 steps each, with every trained model listed and shared on Hugging Face.

FLUX LoRA Training Tutorial: Kohya GUI on Windows and Cloud GPUs
FLUX LoRA training with Kohya GUI, distilled from 104 trainings into ranked configs for 8 GB to 48 GB GPUs, plus Windows and cloud tutorials.

SECourses 3D Render Style FLUX LoRA Published on CivitAI
SECourses 3D Render FLUX LoRA model achieves style consistency; download it on CivitAI and get the full workflow with experiment results on Hugging Face.

Training a Public FLUX Style LoRA: Captions vs No Captions Test
FLUX style LoRA training in public: captions vs no captions on a 114-image 3D render dataset, 500 epochs on 4x A6000 GPUs with Kohya.

SDXL DreamBooth / Fine Tuning: Best Kohya Configs From 6 GB GPUs
SDXL DreamBooth / fine tuning configs for Kohya SS GUI, based on results of 100+ full trainings, train on GPUs with as low as 6 GB VRAM via block swapping.

SDXL Fine Tuning: 20 New OneTrainer Tests, Results and Better Workflow
SDXL fine tuning with OneTrainer on RealVis XL 4: 20 tests compare batch sizes, learning rates, xFormers, regularization images and loss weight functions.

SDXL Training: AdamW Batch Size 30 vs Adafactor Batch Size 1 Comparison
SDXL training test on an RTX A6000: AdamW at batch size 30 vs the SECourses Adafactor config at batch size 1, which gave far better resemblance.

LoRA Training Tips: Learning Rate, Network Rank, Alpha & Batch Size
Excellent tips for LoRA training, slightly improved: how network rank, alpha, optimizer, batch size and precision change the effective learning rate.

Extract LoRA From FLUX & SDXL DreamBooth Models: Tutorial & Comparison
LoRA extraction from FLUX and SDXL DreamBooth models in Kohya, compared with fine tuning and LoRA training; higher ranks, up to 640, give better results.

OneTrainer SDXL Fine-Tuning Best Presets: DreamBooth on 8.5 GB VRAM
OneTrainer Stable Diffusion XL (SDXL) fine-tuning presets and installers; three presets with the same quality, from 8.5 GB VRAM to the fastest 17 GB.

SDXL DreamBooth Captioning Comparison: WD14 vs Kosmos-2 vs Ohwx Man
Image captioning in SDXL fine-tuning / DreamBooth of a single person reduces likeness with almost no benefit, in this OneTrainer test using 10.3 GB VRAM.

Stable Diffusion Regularization Images: 4K Man & Woman Class Dataset
Stable Diffusion regularization images: 5200 hand-checked real man and 5200 woman photos, cropped and sorted by face quality for SDXL and SD training.











