Tensor.Art
Create

ShadowGost V2

CHECKPOINT
Original


Updated:

665

Information - ShadowGost-v2

ShadowGost, this model is the one that I have dedicated more work, I really liked the publications on Twitter about characters or landscapes that had a good management of lighting and shadows, I always had problems with generating a good composition so I ended up creating this model.

If you want to see more results or know about future models that I will be publishing you can visit me in my social networks, it will help me a lot and you can buy me a coffee to support this content.

Recommendations for use

After several attempts, (I still need to learn more) you can improve the models to make it easier to generate beautiful images.

Shadows and Highlights Handling.

The mainstay of all this is compositing a good prompt (or that's what worked for me).

  • First I detail the quality with tags like (Masterpiece, extremely detailed fine touch).

  • Second, the shadows these tags generate a good result (hard light, studio light, light rays, dappled light, reflection, shadows, ray tracing) depending on the environment you want to develop you can use (warm light, sunlight, natural light).

  • And finally, character development, this is at the convenience of each user, although I always add this tag for its versatility (dynamic pose).

It is recommended to use Hiresfix, if you want instant results, but without it also generates beautiful works.

Image tips:

  • If you are going to generate images with a large resolution or the character you develop is far away I recommend you to use Hires.fix, to improve the quality.

  • Addetailer is a good option if you don't use Hires.fix and improves the details of your image.

  • If you want to get a brush or watercolor painting effect you can use this setting with Hires.fix:

    • Upscaler: Latent

    • Hirest steps: 15 - 20

    • Denoising strength: 0.46 - 0.5

Recommended parameters:

  • Sampler: DPM++ SDE Karras: 20 to 30 steps.

  • Sampler: DPM++ 2M Karras: 20 to 60 steps.

  • Sampler: Euler a: 40 to 60 steps.

  • CFG Scale: 4 to 11.

  • Resolutions: 512x768, 512x1024 for Portrait!

  • Resolutions: 768x512, 1024x512, 1536x512 for Landscape!

  • Hires.fix: Latent, Lanczos, R-ESRGAN 4x+Anime6b.

  • Clip Skip: 2.

  • Negatives: ' (worst quality, low quality:1.4), (zombie, sketch, interlocked fingers, comic) '

  • Negatives if you can't use Hires.fix: '(worst quality:1.6, low quality:1.6), (zombie, sketch, interlocked fingers, comic)'

Version Detail

SD 1.5
Images Generated by kmisogi

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.