LORA Photorealism for FLUX

This repository provides a checkpoint with trained LoRA for FLUX.1-dev model by Black Forest Labs

ComfyUI

Download the model and use it in ComfyUI with the FLUX workflow.

Trigger Word

The trigger word is: TOK

Training Details

LoRA for FLUX.1-dev model trained on Replicate.

Inference

from diffusers import FluxPipeline
import torch

# Load the main pipeline
pipe = FluxPipeline.from_pretrained(
    "black-forest-labs/FLUX.1-dev",
    torch_dtype=torch.float16
).to("cuda")

# Load LoRA weights
pipe.load_lora_weights("BartoUno/flux-lora-trained", weight_name="lora.safetensors")

# Generate image
prompt = "TOK, portrait photography, professional lighting"
image = pipe(
    prompt,
    guidance_scale=3.5,
    num_inference_steps=28,
).images[0]

License

lora.safetensors falls under the FLUX.1 [dev] Non-Commercial License