Update comfy UI and download the models bellow from the manager just search for Cascade I leave the size pipelines at the end of the article
My extension
https://github.com/if-ai/IF_prompt_MKR
Cascade extension A1111 -Forge
https://github.com/blue-pen5805/sdweb-easy-stablecascade-diffusers
ollama
https://ollama.com/blog/windows-preview
https://github.com/ollama/ollama
Ollama WEBUI
https://github.com/ollama-webui/ollama-webui
ollama Models
https://ollama.com/impactframes/mistral_alpha_xs
https://ollama.com/impactframes/stable_diffusion_prompt_maker
https://ollama.com/brxce/stable-diffusion-prompt-generator
Satable-Cascade
https://huggingface.co/stabilityai/stable-cascade
ProteusV0.3
https://huggingface.co/dataautogpt3/ProteusV0.3
ConfyUI WF
https://gist.github.com/comfyanonymous/0f09119a342d0dd825bb2d99d19b781c#file-stable_cascade_workflow_test-json
Discord Banodoco
https://discord.gg/7UyNtVXB
VRAM for ComfyUI give or take
full
stage_a.safetensors, stage_b.safetensors, stage_c.safetensors and text encoder model.safetensors = 20GB+
bfp16
stage_a.safetensors, stage_b_bf16.safetensors, stage_c_bf16.safetensors and text encoder model.safetensors = 14GB
lite
stage_a.safetensors, stage_b_lite.safetensors, stage_c_lite.safetensors and text encoder model.safetensors = 8GB
lite_bf16
stage_a.safetensors, stage_b_lite_bf16.safetensors, stage_c_lite_bf16.safetensors, and text encoder model.safetensors = 6GB
SD A1111 and Forge extension
16GB VRAM