Summary
Keywords
Full Transcript
In this episode, I’ll show you how to train a Flux Kontext LoRA using Fal AI, including step-by-step image prep, training settings, and how to make it work with both the standard Flux Kontext and the Nunchaku version in ComfyUI. We’ll cover: - How to create proper before/after image pairs - Setting up trigger words and captions - Exporting LoRAs in Comfy format - Converting LoRAs for Nunchaku compatibility - Using Power Lora Loader and Nunchaku Lora Loader - Fixing common errors like UNet final layer issues - Creating sketch-to-icon workflows with consistent alignment - Testing different seeds and prompt tweaks for better output - Training LoRAs for characters, fantasy items, and mandalas Get the workflows and instructions from discord for free Step 1: use this link to join if you didn't join already https://discord.gg/gggpkVgBf3 Step 2: access directly this episode using this link: https://discord.com/channels/1245221993746399232/1394631896960335872/1394632385689157724 Fal AI Lora training page for Flux Kontext https://fal.ai/models/fal-ai/flux-kontext-trainer Lora Patch (+ optional bat file from discord for easy conversion and running) https://huggingface.co/lym00/comfyui_nunchaku_lora_patch/tree/main and download this file patch_comfyui_nunchaku_lora.py Check Other Episodes https://www.youtube.com/playlist?list=PL-pohOSaL8P9kLZP8tQ1K1QWdZEgwiBM0 Unlock exclusive perks by joining our channel: https://www.youtube.com/channel/UCmMbwA-s3GZDKVzGZ-kPwaQ/join #comfyui #fluxkontext #comfyuitutorial Chapters: 00:00 Introduction 00:21 Fal AI Flux Kontext Trainer 01:12 Style Lora - Sketch to Fantasy Game Items 03:23 ATENTION for ComfyUI Users 03:36 Style Lora Second Part 05:24 Nunchaku Lora Error Fix 07:20 Style Lora Third Part 10:47 Character or Person Lora 15:04 Transform to Mandala Lora
