You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Eyght SDXL-Turbo LoRA (300-image fine-tune)

A LoRA adapter fine-tuned on stabilityai/sdxl-turbo from a dataset of ~300 images. Use it to steer SDXL-Turbo toward the style/subject of the training set.

Built, trained, and owned by Eyght. Free Hugging Face model repository.

Model details


Base model	`stabilityai/sdxl-turbo`
Type	LoRA adapter (`pytorch_lora_weights.safetensors`)
Task	Text-to-image
Framework	diffusers + PEFT
License	CreativeML OpenRAIL-M

How to use

from diffusers import AutoPipelineForText2Image
import torch
pipe = AutoPipelineForText2Image.from_pretrained("stabilityai/sdxl-turbo", torch_dtype=torch.float16, variant="fp16")
pipe.load_lora_weights("Eyght/my-sdxl-lora-300")
pipe.to("cuda")
pipe("your prompt, highly detailed", num_inference_steps=4, guidance_scale=1.0).images[0].save("out.png")

Training

LoRA fine-tuned on a custom ~300-image dataset. Checkpoints (checkpoint-100..checkpoint-500) and TensorBoard logs are included.

License & attribution

CreativeML OpenRAIL-M. Built and owned by Eyght.

Citation

@misc{eyght_sdxl_lora_300,
  title  = {Eyght SDXL-Turbo LoRA: an Eyght fine-tune of SDXL-Turbo},
  author = {Eyght},
  year   = {2026},
  howpublished = {Hugging Face, https://huggingface.co/Eyght}
}

Built by Eyght with the local Eyght Veta studio.

Downloads last month: 18

Model tree for Eyght/my-sdxl-lora-300

Base model

stabilityai/sdxl-turbo

Adapter

(85)

this model

Eyght
/

my-sdxl-lora-300