The Ultimate ComfyUi Workflow for Architects
This is the Ultimate ComfyUi Workflow for Architects.
Text to Image SD3.5
SD3.5 Medium
Stable_Diffusion_3_5_Medium.safetensors
https://huggingface.co/stabilityai/stable-diffusion-3.5-medium
SD3.5_Checkpoint_Model:
https://huggingface.co/stabilityai/stable-diffusion-3.5-medium/resolve/main/sd3.5_medium.safetensors
huggingface-cli download stabilityai/stable-diffusion-3.5-medium sd3.5_medium.safetensors --local-dir temp
Example Workflow:
https://huggingface.co/stabilityai/stable-diffusion-3.5-medium/resolve/main/SD3.5M_example_workflow.json
huggingface-cli download stabilityai/stable-diffusion-3.5-medium SD3.5M_example_workflow.json --local-dir temp
Hugging Face Section
Load Library: pip install huggingface_hub
Login: huggingface-cli login
Token: Your Token goes here
Example--------------------------------------------------------------
huggingface-cli download (repo id) (file name)
Positive Prompt:
a contemporary farmhouse cabin in the parametric style of Zaha Hadid in the woods with mist and fog in the air, snowy mountainous landscape with lots of large evergreen trees, soft desaturated rose color glow, ethereal heaven like light, ambient occlusion, soft shadows, groups of people scattered in the foreground
Negative Prompt:
old, worn, faded
Image to text ( Describe ):
Goto ComfyUi Manager _ Click on Custom Nodes Manager _ Search for Janus Pro _ #66 by cychenyue
You will need to install the model below through the ComfyUi Manager under Custom Models. You can search for the one below.
deepseek-ai/Janus-Pro-1B
In addition Goto ComfyUi Manager _ Click on Custom Nodes Manager _ Search for ComfyUI-Easy-Use _ Number #22 by yolain
Text to Audio:
https://huggingface.co/stabilityai/stable-audio-open-1.0/tree/main
stable-audio-open-1_0: (put in models/checkpoint) It’s Important to rename file
huggingface-cli download stabilityai/stable-audio-open-1.0 model.safetensors --local-dir temp
https://huggingface.co/google-t5/t5-base/tree/main
t5-base: (put in models/text_encoders) It’s Important to rename file
huggingface-cli download google-t5/t5-base t model.safetensors --local-dir temp
Image to Video:
https://huggingface.co/Lightricks/LTX-Video
ltx-video-2b-v0.9.5: (put in models/checkpoint)
huggingface-cli download Lightricks/LTX-Video ltx-video-2b-v0.9.5.safetensors --local-dir temp
https://huggingface.co/comfyanonymous/flux_text_encoders/tree/main
t5xxl_fp8_e4m3fn_scaled: (put in models/text_encoders)
huggingface-cli download comfyanonymous/flux_text_encoders t5xxl_fp8_e4m3fn_scaled.safetensors --local-dir temp
Custom Node Utilities:
Crystools
Flux - Comfy-Org/flux1-dev (I didn’t like)
huggingface-cli download Comfy-Org/flux1-dev flux1-dev-fp8.safetensors --local-dir temp
Sample Workflows
https://comfyanonymous.github.io/ComfyUI_examples/flux/
Jupyter Notebook Commands (Linux)
cd – change directory
ls – list files in directory
rm – remove
#comfyui #comfyuistablediffusion #stablediffusion #stabledifussion3.5 #ai #generativeai