Modelpublic

hodgy/shantae-krea2-turbo-v1

sourceHugging Faceupdated 1mo agoView on Hugging Face

0likes26downloads

Model Card

Shantae Krea 2 Turbo LoRA v1

Identity LoRA trained for Krea 2 Turbo with Ostris AI Toolkit.

Usage

—Trigger word: shantae
—Base model: krea/Krea-2-Turbo
—LoRA file: `shantae_krea2_turbo_v1_step3000.safetensors`
—Recommended starting point: 9 steps, guidance 1, LoRA strength 0.8–1.0

Example prompt:

text

shantae, close-up portrait, natural window light, detailed skin, shallow depth of field

Sample images

All samples below were generated from the final step-3000 checkpoint at 1024×1024.


[image]	[image]
[image]	[image]
[image]	[image]
[image]	[image]
[image]	[image]

Training details

Setting	Value
Training steps	3,000
LoRA rank / alpha	32 / 32
Precision	BF16
Optimizer	AdamW8Bit
Learning rate	0.0001
Batch size	1
Gradient accumulation	1
Resolution buckets	512, 768, 1024
Scheduler	FlowMatch
Sample steps / guidance	9 / 1

The original AI Toolkit configuration is included as `config.yaml`.

Notes

Generated images may reproduce limitations or biases present in the training data and base model. Review outputs before use, especially in public or commercial contexts.