Tensor.Art
Create

Monody XL

CHECKPOINT
Original


Updated:

856

My first SDXL Models 🎉🎉

This model is made for cute and adorable characters~

Currently this model is still in the Alpha stage, there are still many things that need to be improved, especially the stability of the body's anatomy.

Due to the special dataset used, this model does not fit some LoRas

and making characters without LoRa is also not possible (I've tested it and the results can only imitate 50% of the characters)

Never use embedings from SD1.5. because it doesn't work on SDXL models.

Don't use weird aspect ratios, the dataset used only works with general aspect ratios.

Please don't be too harsh with this model, such as using very long and confusing prompts, it will produce bad results.

Usage recommendation

You can use Hiresfix or not, the results are almost the same, but I recommend using it.

Hiresfix

  • 4x-AnimeSharp

  • R-ESRGAN 4x+ Anime6B

Sampling Method

  • Euler Ancestral

  • DPM++ 2M Karras

Steps : 20+

CFG scale : 7 - 10

Resolution : 768x1152 (Portrait), 1152x768 (Landscape)

Quality Prompt : masterpiece, best quality

Negative Prompt : worst quality, low quality, normal quality, disfigured

Making XL model very burning resources. Hopefully, this model is worth it.

if you like this model, you can support me by buy me a ko-fi

Thank you.

Version Detail

SDXL 1.0
10850
10
Trained A2 with around 1.1k high quality image. Reorganized dataset.

Project Permissions

    Use Permissions

  • Use without crediting me

  • Share merges of this model

  • Use different permissions on merges

    Commercial Use

  • Sell generated images

  • Use on generation services

  • Sell this model or merges

Comments

Related Posts

Describe the image you want to generate, then press Enter to send.