I know Jack shit about Artifical Intelligence…

  • Shift ✕?

    Shift ✕?

    Somebody on Reddit suggested to give Z-Image Turbo a try. While I already used that one to generate Anna and Elsa images (with disappointing results) I didn’t give it a try for more realistic images. When I loaded Z-Image into Invoke, I noticed a slider I hadn’t seen on other models before; “Shift ✕” with…

  • Anna and Elsa!

    Anna and Elsa!

    Yes! People who know me saw this coming already even before I started this blog. Being a massive Disney Frozen fan, my ultimate goal in this whole A.I. exploration adventure, is to create my own Anna and Elsa images… obviously… So, today I am going to try to have some of the starter models recreate…

  • FLUX.1 dev

    FLUX.1 dev

    Another Starter Model in Invoke is FLUX.1 dev. After installing it from the Model Manager, I ended up with 5 models to choose from: As the name already suggests, FLUX Fill is not meant for image generation in itself. When attempting to do that, Invoke will generate an error, suggesting to use one of the…

  • Dreamshaper 8

    Dreamshaper 8

    After a clean restart, I installed only the first of the Starter Model Bundles, which is called Stable Diffusion 1.5. This left me with Dreamshaper 8, which is already a bit confusing to a noob like me. I have no clue how Stable Diffusion and Dreamshaper relate to each other, but I am sure Google…

  • I made a mess… restart…

    I made a mess… restart…

    I spend a few weeks experimenting with Invoke now and changed a lot of things, downloaded a lot of stuff and restored a lot of backups. Being an absolute noob and not knowing what I was doing at all, I made quite a mess on my system with various copies of models, installations distributed over…

  • Anima?

    Anima?

    When taking another look at the “Starter Bundles” section in Invoke, I noticed “Anima.” I am not sure if it was there before, but I am going to have a look at it. The model installed is called Anima Base 1.0, and it defaults to 1024×1024 images. Let’s give it one of my favorite prompts:…

  • Well, that didn’t work!

    Well, that didn’t work!

    I installed a super simple and cheap 3rd graphics card on my computer with the intention of having both RTX cards fully available for AI stuff and use the simple card for my displays. The card is an NVIDIA NVS 310, and it IS displayed in Windows system info; it just didn’t output any video.…

  • Second GPU added

    Second GPU added

    Received my second NVIDIA Quadro RTX 4000 today and installed it in the computer. Yeah, it’s an ancient card, I know, but it’s all I can afford at the moment. Anyhow, I didn’t feel like wasting too much time in finding out how I could distribute the workload among both GPUs; I just wanted to…

  • Welcome to AI Noob Online!

    Welcome to AI Noob Online!

    So, this is my very first post on this brand new website. Yet it will not be the first post you see. That’s because I already started blogging a bit on my main site before I decided to start a completely separate and new site about my “AI adventures,” and I am going to transfer…

  • Inpainting

    Inpainting

    Inpainting is an amazing function to tell the AI to “pick up where it left off.” This is probably a noob description, but it basically means you can extend pictures in any direction you want. Let’s start with a simple image… It kind of failed miserably on the face, but let’s continue like this is…

  • AI Model Comparison

    AI Model Comparison

    Trying out various A.I. models by generating 3 images with the exact same prompt on all of them. To keep sizes down for the test images, I used a 512 x 512 pixel image size, which is the default for many models. The prompt is “Cute petite redhead with freckles in a sexy pink bikini.…

  • Exploring A.I. on a local computer

    Exploring A.I. on a local computer

    Finally decided to install my own LLM’s and image generation AI systems on my home computer because online services are getting priced ridiculously and limited more and more. I decided to start by feeding the image generation A.I. with very simple and short titles of famous hits. These are my first results… Dreamshaper 8 (No…