Type | |
Stats | 15,078 323,986 |
Reviews | (1,716) |
Published | Feb 6, 2024 |
Base Model | |
Training | Steps: 5,260 Epochs: 4 |
Usage Tips | Strength: 1.3 |
Trigger Words | dark chiaroscuro low-key |
Hash | AutoV2 490E897006 |
Please consider joining my Patreon so I can keep most of my work available for everyone - I'll be releasing early access models there for a cheaper monthly fee than buying them here individually, and also will be providing exclusive models and more!
For business inquiries, commercial licensing, custom models, and consultations, please get in touch at [email protected] or [email protected]. You can also contact me here through CivitAI DM or join my Discord.
Introduction
This LoRA is specifically designed to enhance dark scenes, injecting them with atmospheric depth. Unlike many existing SDXL models which often render dark scenes with a stark, artificial effect, resembling a staged model shoot rather than realistic environments, this LoRA model addresses this limitation. By preserving tonal range in dark scenes more effectively, it brings out subtle details in shadows while maintaining a rich spectrum of tones. This not only creates a more immersive atmosphere but also provides greater flexibility for contrast adjustments in post-processing, eliminating the issue of a too strong black-level or a too spotlight filled scene.
Triggers
dark, chiaroscuro, low-key
In general it works without trigger words fine as well (intended), but use it solely for scenes intended to be dark. You'll find that you're going to use the word 'dark' anyway :). Reminder, it will not work very well for high-key lighting scenes, since you're in essence fighting the LoRA when you do.
Recommended settings
Strength between 0.6 and 2.