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

Base Blend: Built on the Style A2B8 (20% Style A, 80% Style B) FeMix blend strategy. This blend strongly emphasizes the unique and experimental style of Style B.
HassakuXL v2.1fix CLIP Integration: Leverages the improved CLIP (Contrastive Language-Image Pre-training) from HassakuXL v2.1fix. You may observe more consistent adherence to prompt instructions compared to v2.1, resulting in something more similar to the prompt.
HassakuXL v2.1fix VAE Integration: Integrates the VAE (Variational Autoencoder) from HassakuXL v2.1fix. Artifacts should be better managed due to higher style fidelity due to this feature, making it a refined version and less chaotic with the art.
Style B's Style Is Still Implemented: The key element is that the final product will have much more influence of Style B to provide the great effects that are known by the community to be well received.
High Detail Results For Great Value: By understanding what the key details are, you can work in conjunction with the model's output to focus on the better benefits.
Not Reliable With LoRA Use: Make sure to use LoRAs at your own discretion, as it is very unlikely to be producing results that are to your expectations.
For Expert Use: It is something best for an expert, where they understand on the key details to adjust so that the end result is something that looks great.
Post-Process is Strongly Recommended: To fix elements and add more to the product.

In summary: With the balance of A and B's combination, it makes an impressive look for all the products. As a unique balance, it enables people who are great at making models an output to show their prowess by creating new styles and pushing the limitations of what can be achieved.