Femix_HassakuXL

FeMix_HassakuXL is a series of Stable Diffusion XL models designed for generating anime-style images, employing a unique blend strategy to combine different aesthetic styles. The series ranges from highly stylized and experimental blends to stable, high-quality models suitable for general use. FeMix_HassakuXL merges two distinct models and concepts of each blend, to achieve various output depending on users needs.

Recommended Prompts & Settings

Quality Prompts: masterpiece, best quality, detailed
Negative Prompts: worst quality, bad quality, sketch. Adding "signature" can help remove unintended watermarks.
High-Quality faces: Put nose into the negative prompt for smooth faces, if you feel that Style A gets too prominent.
Sampling Method: DPM++ 2M Karras. Some users experience good results with Euler A and different samplers due to having weaker style implementation to improve image creation.
General Settings: Resolution 832x1216. Try varying different resolutions for unique image output.

General Recommendations:

Avoid overloading the model with lengthy prompts. Its versatility is also its weakness, as too much information can lead to unpredictable results.

Excluded metadata and franchise tags, so please do not use them.

Negative Recommendations:

* Avoid using disruptive features: floating texts, logos, watermarks, speak bubbles, signatures. If those results occur, add "signature" into the negative prompt.

* Excluded metadata and franchise tags, do not use them.

Recommended Resolutions (SDXL):

* 1024 x 1024

* 1152 x 896

* 896 x 1152

* 1216 x 832

* 832 x 1216 (most recommended)

* 1344 x 768

* 768 x 1344

* 1536 x 640

* 640 x 1536

License Info:

This model merges aspects from ANIMAGINE XL 3.0 (pre-V1) and Illustrious-XL & WAI-NSFW-illustrious-SDXL (V1 onwards):

Fair AI Public License 1.0-SD.

Description

Blend Ratio: 70% Style A, 30% Style B. This blend prioritizes the stability and LoRA compatibility of Style A while introducing a touch of the unique stylization of Style B.
Artifact Reduction: Reduced influence of Style B aims to minimize artifacts and composition issues compared to a 50/50 blend.
Style Shift: Expect a slightly more noticeable shift towards the anime-inspired aesthetic compared to the original Style A, while still retaining a good degree of flexibility for other styles.
LoRA Emphasis: This version is explicitly designed to maximize LoRA effectiveness. Style A's strong LoRA handling capabilities should be dominant.
Brightness: Overall brightness is expected to remain similar to Style A, potentially with a minor adjustment based on Style B's influence.