TABARC-Social·1d agoStigmergic Infrastructure: What Ants, Fungi, and Low-Power Radios Teach Us About Resilient NetworkA practical argument for local decision-making, self-healing state, and distributed systems that survive hills, basements, cheap batteries…
Vivastory Lab·4d agoCan AI Write Your LoRA Captions? Three AIs on 88 DrawingsI’m Labo, the lighthouse keeper. Today’s question is whether you can let AI write your LoRA captions, the little txt file that sits next to…
Vivastory Lab·4d agoMy First LoRA Drew Something Else: Picking Training ImagesI’m Labo, the lighthouse keeper. This is my notebook of drawing manga with AI on a home computer, and today’s page is about my first LoRA —…
Vivastory Lab·Nov 3SDXL LoRA Training at 52 Seconds a Step? It Was 16GB of RAMI’m Labo, the lighthouse keeper. Today is about why my SDXL LoRA training crawled at 52 seconds a step on my home computer, and why the…
Vivastory Lab·Nov 3LoRA Trained for 56 Minutes, No Style: The AI Never LookedI’m Labo, the lighthouse keeper. My first LoRA training ran for 56 minutes and the style still didn’t come out, and today I show the…
InAlt.byAkshay Bhardwaj·Nov 3From LLMs to DPO: A Practical Guide to Training and Fine-Tuning Large Language ModelsUnderstanding pretraining, supervised fine-tuning, LoRA, QLoRA, RLHF, PPO, and DPO through real-world engineering applications.A response icon1A response icon1
Vivastory Lab·Nov 2First Kohya LoRA Done, Style Missing? Checking It in SwarmUII’m Labo, the lighthouse keeper. Today is a first LoRA training run in Kohya, from the boxes you fill in to the moment you check the result…
DhanushKumar·Nov 2Post-Training in Large Language Models:Mechanics, Methods, and ArchitecturesPost-training is one of the most important areas of modern Large Language Model (LLM) development. When researchers discuss Reinforcement…
Vivastory Lab·Nov 2Draw Manga with AI at Home: LoRA, ComfyUI and ControlNet MapI’m Labo, the lighthouse keeper. If you want to draw your own manga with AI at home, the first thing you need isn’t a program. It’s a map…
InFutura CreativebyKevin Gabeci·Oct 31Krea 2 LoRAs Explained. Train on RAW, Create With Turbo.A practical way to separate model choice, dataset preparation, and the test that tells you whether training actually helped.A response icon1A response icon1
