AI workshop

Prompt crafting

Stable Diffusion 1.5

Stable Diffusion 2.1 (sometimes slow/busy)

SDXL + fine-tuned models (choose: sd_xl_base_1.0.safetensors) (released after SDXL) (SDXL – fast image creation)



Stable Horde (free alternative. slow!)
check also Interrogation option for image captions.

Inpainting / generative fill (Outpainting):
Adobe generative fill

Inpainting based on Stable Diffusion

Image Mixer
Image Mixer demo hosted @

images for the workshop
prompt crafting – doc
examples of ComfyUI workflows

ControlNet V1.1 (examples how-to-use)
ControlNet V1.0 (older version)

ComfyUI (first look to ComfyUI) (test example workflows here)
Kohya_ss (training fine tune models) (sources for learning etc.)

KREA-AI (image/video (check demo!))
ZeST – Zero-Shot Material Transfer from a Single Image (image-to-3D)
gen-2 text-2-video
audioldm-text-to-audio-generation (text-to-speech)
virtual avatar – test various image models

Aaron Hertzmann – When Machines Change Art
Lev Manovich & Emanuele Arielli – Artificial Aesthetics chapter 5 (PDF -> “Media Translations”)
Vinoy Laughner – Computer Art: Is It Really Art?

Additional links:
Generative Landscape -course : CLIP etc (Camenduru webui explained and installation instructions, notice min. 4GB GPU required) (diffusion for small datasets)
pix2pixHD next-frame-prediction (used on FATUM)