A pack of nodes(only 2 for now) to make my life easier and hopefully yours ;)
A pack of nodes, to make my life and hopefully your a bit easier.
<img width="1472" alt="image" src="https://github.com/user-attachments/assets/b23b0c69-eda6-49d2-8186-dc842de268d6">Clone the repository:
git clone https://github.com/MzMaXaM/ComfyUi-MzMaXaM.git
ComfyUi Manager
ComfyUi Manager/ Custom Nodes Manager/ Search for "mzmaxam"
A custom node for ComfyUI designed to generate an empty latent image at predefined resolutions. This node is useful for initializing images for further processing.
"Cinema landscape (1536x640) 21x9",
"Smartphone landscape (1472x704) 19x9",
"Hd Wide landscape (1344x712) 17x9", #New
"HD_Video landscape (1280x720) 16x9", #Updated
"Laptop landscape (1216x768) 8x5",
"Photo landscape (1216x832) 3x2",
"Widescreen landscape (1176x840) 7x5",
"TV landscape (1152x896) 4x3",
"Square (1024x1024) 1x1",
"TV portrait (896x1152) 4x3",
"Widescreen portrait (840x1176) 7x5",
"Photo portrait (832x1216) 3x2",
"Laptop portrait (768x1216) 8x5",
"HD_Video portrait (720x1280) 16x9", #Updated
"Hd Wide portrait (712x1344) 17x9", #New
"Smartphone portrait (704x1472) 19x9",
"Cinema portrait (640x1536) 21x9",
A custom node for ComfyUI designed to use with pony models. It have 2 positive and one negative text encoders. 2 positives becouse in pony models I always want to keep the scores separate from the prompt. 2 positive text are concatenated and send over as one conditional.
Added this version for testing I don't think anyone need it if I'll find were to use it Ill update it...
A custom node for ComfyUI designed to generate an empty latent image at predefined resolutions. This node is useful for initializing images for further processing.
"Cinema landscape (2176x960) 21x9",#2.26
"Smartphone landscape (2080x992) 19x9",#2.09
"HD Wide landscape (1904x1008) 17x9",#1.89
"HD Video landscape (1920x1080) 16x9",#1.77
"Laptop landscape (1720x1088) 8x5",#1.58
"Photo landscape (1720x1184) 3x2",#1.45
"Widescreen landscape (1664x1184) 7x5",#1.4
"TV landscape (1632x1280) 4x3",#1.28
"Square (1448x1448) 1x1",#1
"TV portrait (1280x1632) 4x3",#0.78
"Widescreen portrait (1184x1664) 7x5",#0.71
"Photo portrait (1184x1720) 3x2",#0.68
"Laptop portrait (1088x1720) 8x5",#0.63
"HD Video portrait (1080x1920) 16x9",#0.56
"HD Wide portrait (1008x1904) 17x9",#0.53
"Smartphone portrait (992x2080) 19x9",#0.48
"Cinema portrait (960x2176) 21x9",#0.44
I created this node for my workflows in which I'm using 16:9 resolution, and from (1280x720) 16x9, it's exactly 1.5 times smaller than (1080x1920) 16x9
That's why I made it :D
Worked on a node that would send the input image to the local Ollama Models. You have to Install Ollama and add Visual Models for it to work. As of today I can recommend minicpm Model Bear in mind that the LLM model will eat resources...
As the name suggests this node should generate a prompt for you from the prompt you'll give it :D Don't forget that LLM models eat your resources.