import torch
from diffusers import DiffusionPipeline
from diffusers.utils import load_image
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("black-forest-labs/FLUX.2-klein-9B", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("Leon1000/Flux2-Klein-9B-Consistency")
prompt = "Turn this cat into a dog"
input_image = load_image("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/cat.png")
image = pipe(image=input_image, prompt=prompt).images[0]Run online:https://www.runninghub.ai/post/2028302502973677570?inviteCode=rh-v1331
Welcome to join our Discord channel for discussion, or contact me to collaborate on custom LoRA builds: https://discord.gg/yVAVa43mWk
LoRA can significantly improve Klein consistency without any cue words. Video tutorial: https://youtu.be/JXMbbbdfnSg
----- 2026/4/17 V2 version update ----
This time, I systematically break down how to train the Klein model, including dataset creation strategies and a detailed training tutorial: https://youtu.be/j6dqOekUQ8c
Cloud training — sign up to get two free 3-hour RTX 5090 vouchers: https://studio.aigate.cc/images/993593021914284032?channel=R6P1L7N3J
Significantly resolved the color cast issue.
Fixed the issue of images appearing dirty due to excessive detail additions in V1.
Reduced the issue of overly high saturation in images generated by Klein.
- Downloads last month
- 13
Model tree for Leon1000/Flux2-Klein-9B-Consistency
Base model
black-forest-labs/FLUX.2-klein-9B