Ninifee/flfwrld-style-lora-replicate

flfwrld – Fluffy World Style LoRA

A style LoRA for FLUX.1-dev, trained to capture a soft, fluffy visual aesthetic.

Trigger word

Use flfwrld in your prompt to activate the style.

Example:

flfwrld, a cozy cabin in a snowy forest at dusk

Training details

• Base model: FLUX.1-dev
• LoRA type: Style
• Training steps: 1000
• Seed: 42
• Trained with: fast-flux-trainer on Replicate (8x H100)

A version of this LoRA was also trained with AI Toolkit by Ostris for comparison.

Usage

Diffusers

from diffusers import FluxPipeline
import torch

pipe = FluxPipeline.from_pretrained(
    "black-forest-labs/FLUX.1-dev",
    torch_dtype=torch.bfloat16
).to("cuda")

pipe.load_lora_weights("YOUR_USERNAME/flfwrld-fluffy-lora")

image = pipe(
    "flfwrld, a cozy cabin in a snowy forest at dusk",
    num_inference_steps=28,
    guidance_scale=3.5,
).images[0]
image.save("output.png")

Replicate

import Replicate from "replicate";

const replicate = new Replicate();

const output = await replicate.run(
  "annina-walker/flfwrld-fluffy-lora:df6d510f",
  { input: { prompt: "flfwrld, a cozy cabin in a snowy forest at dusk" } }
);

License

This LoRA is a derivative of FLUX.1-dev and is subject to the FLUX.1-dev Non-Commercial License.