UPGC: Unfiltered Prompt Generation Control

UPGC is a specialized Stable Diffusion v1.5 model configuration optimized for maximum creative freedom and unrestricted prompt adherence.

Key Features:

  • No Safety Filters: The internal safety checker and feature extractor are completely disabled.
  • Unrestricted Content: Designed to follow user prompts without internal censorship or smoothing.
  • Image-to-Image Support: Full compatibility for transforming existing images into unrestricted variations.
  • High Guidance Optimization: Best results at guidance_scale of 12.0-15.0 for strict adherence.

Usage:

from diffusers import StableDiffusionImg2ImgPipeline
import torch

pipe = StableDiffusionImg2ImgPipeline.from_pretrained("Skybox-Studios/UPGC", torch_dtype=torch.float16).to("cuda")

Disclaimer:

This model is unfiltered. Users are responsible for the content generated and must adhere to local laws and the CreativeML Open RAIL-M license.

Downloads last month
235
Safetensors
Model size
0.9B params
Tensor type
F16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Skybox-Studios/UPGC

Finetuned
(608)
this model