metadata

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
language:
  - en
tags:
  - flux
  - diffusers
  - lora
  - replicate
base_model: black-forest-labs/FLUX.1-dev
pipeline_tag: text-to-image
instance_prompt: Retro TOK illustration style of
widget:
  - text: >-
      Retro TOK illustration style of a lone hiker standing triumphantly on the
      summit of a snow-capped mountain. The hiker, wearing a red jacket and
      backpack, is silhouetted against a clear blue sky. The word "ASCEND" is
      painted in bold, blue letters at the base of the mountain, adding a
      powerful message to the image. The overall impression is one of
      determination, achievement, and the beauty of nature. Dramatic high
      quality
    output:
      url: images/ascend.webp
  - text: >-
      Retro TOK illustration style of a vivid aurora borealis lighting up a
      landscape with norwegian houses
    output:
      url: images/aurora.webp
  - text: >-
      Retro TOK illustration style of a cyclist riding through a scenic mountain
      trail
    output:
      url: images/bike.webp
  - text: >-
      Retro TOK illustration style of a person giving a ted talk on a TED stage
      with the TED logo, "the speaker"
    output:
      url: images/ted.webp
  - text: >-
      Retro TOK illustration style of a quaint, old-fashioned bookstore, with
      the words "Tales of Wonder" etched in gold lettering above the storefront
    output:
      url: images/bookstore.webp
  - text: >-
      Retro TOK illustration style of a painting of a barista creating an
      intricate latte art design, with the "Coffee Creations" logo skillfully
      formed within the latte foam, grey linedrawing background
    output:
      url: images/coffee.webp
  - text: >-
      Retro TOK illustration style of a painting of a barista creating an
      intricate latte art design, with the "Coffee Creations" logo skillfully
      formed within the latte foam
    output:
      url: images/coffee2.webp

Prompt
Retro TOK illustration style of a lone hiker standing triumphantly on the summit of a snow-capped mountain. The hiker, wearing a red jacket and backpack, is silhouetted against a clear blue sky. The word "ASCEND" is painted in bold, blue letters at the base of the mountain, adding a powerful message to the image. The overall impression is one of determination, achievement, and the beauty of nature. Dramatic high quality

Prompt
Retro TOK illustration style of a vivid aurora borealis lighting up a landscape with norwegian houses

Prompt
Retro TOK illustration style of a cyclist riding through a scenic mountain trail

Prompt
Retro TOK illustration style of a person giving a ted talk on a TED stage with the TED logo, "the speaker"

Prompt
Retro TOK illustration style of a quaint, old-fashioned bookstore, with the words "Tales of Wonder" etched in gold lettering above the storefront

Prompt
Retro TOK illustration style of a painting of a barista creating an intricate latte art design, with the "Coffee Creations" logo skillfully formed within the latte foam, grey linedrawing background

Template:

Retro TOK illustration style of

If you want to enforce the grey linedrawing of the background a bit more, include the words grey linedrawing background

Flux_Lora_Retro_Linedrawing_Style_V1

Dataset based on manually processed images from: Dictionnaire universel d'histoire naturelle https://www.biodiversitylibrary.org/page/19537981

This only used the images from the book above. There are more different volumes with different subjects.

Trained on Replicate using:

https://replicate.com/ostris/flux-dev-lora-trainer/train

Trigger words

You should use TOK to trigger the image generation.

Use it with the 🧨 diffusers library

from diffusers import AutoPipelineForText2Image
import torch

pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.float16).to('cuda')
pipeline.load_lora_weights('SebastianBodza/flux_lora_retro_linedrawing_style_v1', weight_name='lora.safetensors')
image = pipeline('your prompt').images[0]

For more details, including weighting, merging and fusing LoRAs, check the documentation on loading LoRAs in diffusers