AETHER DIVINE AEGIS

LORA
Original


Updated:

The AETHER DIVINE AEGIS LoRA is a powerful aesthetic model designed to transport your creations into a realm of opulent, high-fantasy divinity. Based on a curated set of images featuring majestic gods, celestial warriors, and supreme royalty, this LoRA specializes in generating stunning, hyper-realistic art with a defined style of luxury and power.

The core aesthetic is defined by a dramatic, high-contrast palette of Black, Gold, and White/Cream, accentuated by luminous golden light effects. Expect intricate, highly detailed character designs featuring filigreed Gold Armor, regal attire, glowing weapons, and an overall aura of immense, divine authority.

Suggested Generation Parameters

To best utilize the high detail and strong contrasts inherent in the AETHER DIVINE AEGIS style, use the following suggested parameters. These settings are optimized for sharp detail, rich coloring, and effective handling of the dramatic lighting and metallic surfaces.

Sampling Method ➡️ DPM++ SDE Karras or Euler A

SDE Karras is excellent for high detail and a painted look. Euler A is a fast, reliable alternative for general use.

Sampling Steps ➡️ 30 - 45

Sufficient steps to render the complex armor, filigree, and hyperdetails without over-baking the image.

CFG Scale ➡️ 6.5 - 9.0

A slightly higher scale is needed to enforce the strong stylistic elements (Black/Gold/Glow) and the high level of detail.

Clip Skip ➡️ 2

Often ideal for stylistic LoRAs, as it can bring out more detail and color depth.

Upscaler ➡️ Latent (Nearest, None, or 2x) for Hires Fix (or) ESRGAN models (like 4x-UltraSharp) for post-generation.

Latent is fast and effective for Hires Fix. Use a sharpening upscaler for final output.

Hires Steps ➡️ 15 - 25

Balances quality enhancement with generation time during the Hires Fix process.

Denoising Strength ➡️ 0.5 - 0.65

A mid-range value is best for Hires Fix; it cleans up imperfections while preserving the core composition and the intended style.

Resolution Size ➡️ 512x768 or 768x512 (Portrait/Full Body)

Start with these standard resolutions and use Hires Fix to double (e.g., to 1024x1536) for maximum detail.

LoRA Weight : Start with a weight of 0.8 to 1.0 to ensure the distinct style is strongly applied. You can lower the weight to blend with other LoRAs or base models.

Version Detail

Trained by Tensor
FLUX.1
700
4

Project Permissions

Reprinting is strictly prohibited

    Use Permissions

  • Use in TENSOR Online

  • As a online training base model on TENSOR

  • Use without crediting me

  • Share merges of this model

  • Use different permissions on merges

    Commercial Use

  • Sell generated contents

  • Use on generation services

  • Sell this model or merges

Related Posts