tsiendragon/qwen-image-finetune
Repo for Qwen Image Finetune
Supports efficient fine-tuning of multiple image editing models (FLUX Kontext, DreamOmni2, Qwen-Image-Edit variants) through LoRA with an optimized embedding cache achieving 2-3x training acceleration. Features multi-resolution mixed training that dynamically selects optimal resolution per image aspect ratio, multi-image composition support, and unified logging across TensorBoard, Weights & Biases, and SwanLab. Built on PyTorch with FSDP distributed training support and comprehensive validation visualization to track model improvements across training steps.
218 stars.
Stars
218
Forks
23
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Mar 12, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/tsiendragon/qwen-image-finetune"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related models
KohakuBlueleaf/LyCORIS
Lora beYond Conventional methods, Other Rank adaptation Implementations for Stable diffusion.
PRITHIVSAKTHIUR/Qwen-Image-Edit-2511-LoRAs-Fast-Lazy-Load
Demonstration for the Qwen-Image-Edit-2511 model with lazy-loaded LoRA adapters for advanced...
Akegarasu/lora-scripts
SD-Trainer. LoRA & Dreambooth training scripts & GUI use kohya-ss's trainer, for diffusion model.
cloneofsimo/lora
Using Low-rank adaptation to quickly fine-tune diffusion models.
LeslieZhoa/Simple-Lora
diffusion lora chinese tutorial，虚拟idol训练中文教程