
How to Train a Wan 2.2 LoRA With musubi-tuner (High and Low Noise, 16 GB and 24 GB)
Train your own Wan 2.2 LoRA locally with musubi-tuner: model files, install, dataset and TOML, caching, high and low noise commands, VRAM settings and ComfyUI use.

Qwen-Image 2.1 Remove-Object LoRA in ComfyUI: Erase Anything in 6-8 Steps
Use Pruna’s Remove-Object LoRA for Qwen-Image 2.1 in ComfyUI: files and folders, the Precise Removal Outline node, masks, settings, limits and fixes.

How to Use Flux in ComfyUI (2026): Which Model, Files, Settings, LoRAs and Fixes
Run Flux locally in ComfyUI: pick FLUX.1, FLUX.2 or FLUX.3 Klein, put each file in the right folder, use the best settings, add LoRAs and fix errors.

How to Train a LoRA With Kohya_ss GUI: Install, Dataset, Settings and Fixes (2026)
Train your own LoRA with the free Kohya_ss GUI: Windows install with uv, the REPEATS_trigger class folder format, captioning, safe starting settings for SD 1.5, SDXL and Flux,…

How to Use LoRAs in ComfyUI: Load LoRA, Stacking, Strength and Fixes (2026)
Use LoRAs in ComfyUI step by step: the models/loras folder, wiring the Load LoRA node, trigger words, strength settings, stacking multiple LoRAs, LoraLoaderModelOnly for Flux, Wan and Qwen-Image,…

MOSS-Image Control LoRAs in ComfyUI: Pose, Canny, Depth and 4 More (Files, Settings)
Use the 7 new open control LoRAs for MOSS-Image-SFT in ComfyUI: pose, canny, depth, normal, lineart, scribble and segmentation. Files, folders, preprocessors, settings and fixes.

MOSS-Image Guide (2026): Files, Settings, LoRAs, Editing and Training
The complete MOSS-Image roadmap for ComfyUI: exact files and folders, VRAM, settings, 4/8-step LoRAs, Kromoss, photo editing, layouts, LoRA training and fixes.

P-H3-Image 2-Step LoRA in ComfyUI: pDMD Setup, Settings, VRAM and 2K Refine (2026)
Cut P-H3-Image from 8 steps to 2 with the pDMD 2-NFE LoRA. Exact file, folder, the four ComfyUI node changes, step advice, VRAM and fixes.

Fizgig 9.2: Train a MOSS-Image LoRA in 20 Minutes (Install, Settings, VRAM and ComfyUI)
Fizgig 9.2 cuts MOSS-Image LoRA training from 49 to about 20 minutes and batch-3 memory from 21.7 GB to 7.1 GB. What changed, supported models and VRAM, install…

ZenTrainer Guide: Train LoRAs Inside ComfyUI With Live Samples (2026)
Install ZenTrainer and train LoRAs for MOSS Image, Krea 2, Z-Image, Qwen-Image, SDXL and MiniMax H3 inside ComfyUI, with live samples, captions and fixes.