wanghaofan's picture
Update README.md
77ff92b verified
|
raw
history blame
2.36 kB
metadata
tags:
  - text-to-image
  - stable-diffusion
  - lora
  - diffusers
  - image-generation
  - flux
  - safetensors
widget:
  - text: logo,Minimalist,Make a picture of a camera with the word 'Me'
    output:
      url: images/73e7db6a33550d05836ce285549de60075d05373c7b0660d631dac33.jpg
  - text: wablogo, Minimalist, banboo and panda,logo
    output:
      url: images/ddcef76ded3c757896813f73ab69525785ac5a3a46e7edad99738726.jpg
  - text: wablogo, Minimalist, Leaf and cat,logo
    output:
      url: images/13a3ce0a23c2059e8d8ac23d61cd577ef6d8fe3d998e4bd92cf140ff.jpg
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: wablogo, logo, Minimalist
license: other
license_name: flux-1-dev-non-commercial-license
license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md

FLUX.1-dev-LoRA-Logo-Design

This is a LoRA (Logo-Design) trained on FLUX.1-dev by CJim on Shakker AI.

Showcases

Prompt
logo,Minimalist,A man stands in front of a door,his shadow forming the word "A",
Prompt
logo,Minimalist,A pair of chopsticks and a bowl of rice with the word "Lee",
Prompt
wablogo,Minimalist,Leaf and cat,logo,

Trigger words

You should use wablogo, logo, Minimalist as trigger words. The recommended scale is 0.8 in diffusers.

Inference

import torch
from diffusers import FluxPipeline

pipe = FluxPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", torch_dtype=torch.bfloat16)
pipe.load_lora_weights("Shakker-Labs/FLUX.1-dev-LoRA-Logo-Design", weight_name="FLUX-dev-lora-Logo-Design.safetensors")
pipe.fuse_lora(lora_scale=0.8)
pipe.to("cuda")

prompt = "logo,Minimalist,A bunch of grapes and a wine glass"
image = pipe(prompt, 
             num_inference_steps=24, 
             guidance_scale=3.5,
            ).images[0]
image.save(f"example.png")

Online Inference

You can also download this model at Shakker AI, where we provide an online interface to generate images.

Acknowledgements

This model is trained by our copyrighted users CJim. We release this model under permissions. The model follows flux-1-dev-non-commercial-license.