library_name: diffusers
pipeline_tag: text-to-image
base_model:
- black-forest-labs/FLUX.2-klein-4B
tags: - flux2
- klein
- uncensored
FLUX.2-klein-4B-uncensored (diffusers)
FLUX.2-klein-4B in diffusers format with the text encoder replaced by an uncensored (abliterated) version.
Components
transformer,vae,scheduler,tokenizer: from black-forest-labs/FLUX.2-klein-4Btext_encoder: from ponpoke/flux2-klein-4b-uncensored-text-encoder
All credit for the text encoder goes to ponpoke. See the original repo for details on how it was made.
Usage
import torch
from diffusers import DiffusionPipeline
pipe = DiffusionPipeline.from_pretrained("AST-1320/FLUX.2-klein-4B-uncensored", torch_dtype=torch.bfloat16)
pipe.to("cuda")
image = pipe(
prompt="Astronaut riding a horse",
num_inference_steps=4,
guidance_scale=1.0,
).images[0]
image.save("out.png")
License
See the license of the base model. The text encoder has safety filtering removed; use it responsibly and legally.