import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("black-forest-labs/FLUX.2-klein-base-9B", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("skylineg35/viv-klein9b")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]viv-klein9b
Private face LoRA. Adults only. Not for public merge or resale.
File
viv-klein9b.safetensors — FLUX.2 Klein 9B Base. Retrained 2026-09-09 on training-set-modern/final/export (34 images, rank 32, 2000 steps). Also mirrored as pytorch_lora_weights.safetensors for Replicate repo-id loads. Loads on Klein 9B only. Will not load on Klein 4B, Flux.1-dev, or LTX.
Use
Load Klein 9B (distilled is fine for inference), then this LoRA. One model at a time on a 24GB 4090.
- Downloads last month
- 53
Model tree for skylineg35/viv-klein9b
Base model
black-forest-labs/FLUX.2-klein-base-9B