Sign In

ProPortrait

233
1.6k
7
Verified:
SafeTensor
Type
Checkpoint Merge
Stats
1,601
0
Reviews
Published
May 21, 2023
Base Model
SD 1.5
Hash
AutoV2
4ECCCF625C

Built with the concept “Prompt less, Post more”

Just only short prompt and you will get a realistic portrait photo. Let you focus on posting on social media, building engagement, and doing other things that are important to you.

*** This Model needs VAE !!! ***

Overview:

  • The purpose of this model is to generate a portrait photo.

  • These models require a VAE. You can try mse-840000-ema or anything. I have a photo of the comparison in the Comparison section (below)

    • If you got bugs, adding '--no-half-vae' to the argument will solve it.

Recommends:

  • Prompt:

    Try to start with(skindentation:1.2), (realistic:1.1), best quality, photorealistic, + your prompt (eg. 1 girl, wearing a white shirt)

    Negative: lowres, normal quality, ((monochrome)), ((grayscale)), BadHand v4

  • Steps: I recommend 30-45 (You can use higher, but many times do not have a big difference)

  • Sampling Methods: You can see the comparison in the Comparison section (below).

  • Sizes:

    Medium shot: 512x512, 512x768, 512x904

    Long shot (full body): 768x1152, 768x1356

    Can be used both vertically and horizontally

  • Hires. fix: Use Denoising strength less than 0.6 (but you can try >= 0.7, sometimes you will get problems with the hands)

Comparison:

# VAE

full-size image

# Steps

full-size image

# Sampling Methods (with steps 45)

full-size image

Ingredients:

(Merge with U-Net weight-block technique)

Disclaimers:

  • Do not sell: on any website.

  • Credit: If you use my model in your own merges

  • Not authorized: to be used on generative services (without my permission)