Model Information

Amalgam_Mix - Image 1
Amalgam_Mix - Image 2

Description

!caution

This model assumes further block merge after download. Using it immediately after download will not produce the desired results.

This text has been translated by deepL. Please excuse the imperfections.

This is a mix model for block merge, invented based on WarriorMama777's report https://huggingface.co/WarriorMama777/OrangeMixs.

It is based on Basil_mix and blended with instagram-latest-plus-clip-v6e1_50000 and Shirayuki_Anime_V1.

It can be used with many anime models. Gives a realistic look and sparkle to the eyes and improves the fingers somewhat.

Depending on the prompt, it can make it look like a screenshot from an open world game!

To use, block merge this Amalgam model with the following distribution for model A of your choice.

  • model A  Some model of your choice

  • model B  Amalgam.safetensors

  • weight   1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0.1,0.3,0.5,0.7,0.9,1

  • base_alpha 1

I prefer to use this for Baka-Diffusion V1. This is the model used in the sample image.

If you find it too realistic, please block merge this Amalgam model with the following distribution instead. This should leave you with a more animated representation.

  • model A Some model of your choice

  • model B Amalgam.safetensors

  • weight 0,0.1,0.3,0.5,0.7,0.9,1,1,1,1,1,1,0,0,0,0,0,0,0,0.1,0.2,0.25,0.2,0.1,0

  • base_alpha 1

It is recommended to use Highres fix and DPM++ 2M karras when using the created model.

Enjoy!

Amalgam_Mix

690downloads
Download Model

Model Details

Type
AI Model
Task
text-to-image
Subtype
Safetensors / Checkpoint AI Model
Created
Updated
June 6, 2026

Versions

Related Models