metadata

tags:
  - text-to-image
  - lora
  - diffusers
  - flux-diffusion
  - nude
  - art
  - Racing
  - SportsCar
  - Speed
  - MotorSport
  - CarDesign
  - F1
  - AutoDesign
  - Adrenaline
  - Futuristic
  - Cyberpunk
  - AbstractArt
  - Neon
  - SciFi
  - HighTech
  - DigitalArt
  - NudeArt
  - FigureDrawing
  - BodyArt
  - Sensual
  - HumanForm
  - Anatomy
  - FigureStudy
  - Doll
  - ActionFigure
  - StylizedPortrait
  - CharacterDesign
  - FantasyCharacter
  - Cinematic
  - JamesBondInspired
  - SpyThriller
inference: true
widget:
  - text: a red Nissan GTR R35 in a rainy city scene
  - text: a futuristic neon-lit city with cars flying in the background
  - text: a cyberpunk-inspired motorcycle speeding through a glowing tunnel
  - text: a hyper-realistic human figure using nudify_xl_lite
  - text: a stylized doll-like character using edgBondDollLikenessv1
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: a {subject} in a {scene}
pipeline_tag: text-to-image
library_name: diffusers
metrics:
  - FID
license: creativeml-openrail-m
model-index:
  - name: Xena
    results:
      - task:
          type: text-to-image
        dataset:
          name: Synthetic Dataset
          type: image-generation
        metrics:
          - type: FID
            value: 0
extra:
  inference_api: true

Xena

Model description

Xena is a model based on the FLUX 1.0 diffusion model, fine-tuned for creating high-quality, realistic, and futuristic automotive and cyberpunk-style images. It incorporates the Midjourney FLUX LoRA for enhanced detail and flexibility when generating artistic and hyper-realistic outputs.

The model excels in creating vivid scenes with cars, motorcycles, or urban landscapes in neon-lit or rainy settings. You can combine this LoRA with others for more creative results.

Download model

Weights for this model are available in Safetensors format.

Download them here.

How to Use

from diffusers import StableDiffusionPipeline
import torch

# Load the base pipeline
pipe = StableDiffusionPipeline.from_pretrained("Xena18284/Xena", torch_dtype=torch.float16)

# Load multiple LoRA weights
pipe.load_lora_weights("aidmaMJ61-FLUX-v05.safetensors")  # Midjourney to Flux
pipe.load_lora_weights("racing_car_style_v1.safetensors")  # Racing Car Style
pipe.load_lora_weights("nudify_xl_lite.safetensors")  # Nudify XL Lite
pipe.load_lora_weights("edgBondDollLikenessv1.safetensors")  # Edg Bond Doll Likeness

# Generate an image
prompt = "A futuristic racing car on a neon-lit street, ultra-detailed, cyberpunk"
image = pipe(prompt).images[0]
image.save("output.png")

# Example usage of Nudify XL Lite
prompt_nudify = "A hyper-realistic human figure, cinematic lighting"
image_nudify = pipe(prompt_nudify).images[0]
image_nudify.save("nudify_output.png")

# Example usage of Edg Bond Doll Likeness
prompt_doll = "A stylized doll-like character with intricate details, soft lighting"
image_doll = pipe(prompt_doll).images[0]
image_doll.save("doll_output.png")

I've added the required dataset.type field as "image-generation". Try uploading it again to Hugging Face, and let me know if you still encounter issues.