README.md

---
tags:
  - text-to-image
  - lora
  - diffusers
  - template:diffusion-lora
widget:
- text: cinestyle retro futuristic car in a cyberpunk city with many lights
  output:
    url: images/2025-02-21_04-24-32_9601.png
- text: cinestyle city with many lights
  output:
    url: images/2025-02-21_04-48-43_4230.png
- text: >-
    night time cinestyle cinematic still: view in messy appartement, woman
    wearing summer dress standing at a window looking at a futuristic dystopia
    city with giant glass pyramides with red tops
  parameters:
    negative_prompt: >-
      anime, cartoon, graphic, text, painting, crayon, graphite, abstract,
      glitch, deformed, mutated, ugly, disfigured
  output:
    url: images/2025-02-21_09-03-31_5619.png
- text: night time cinestyle cinematic still, view on back of a man, a lot of lights
  output:
    url: images/2025-02-21_04-44-07_9716.png
- text: >-
    cinestyle image of a man on a balcony looking at a futuristic city with many
    lights and neon signs
  output:
    url: images/2025-02-21_08-37-04_5779.png
- text: >-
    night time cinestyle cinematic still: Double Exposure photo, featuring eye,
    shattered mirror effect, broken mirror, sharp mirror edges
  output:
    url: images/2025-02-21_09-14-30_6648.png
- text: >-
    night time cinestyle cinematic still: old abandoned robot, begging in
    cyberpunk street corner, coins on the floor
  output:
    url: images/2025-02-21_09-18-17_7465.png
base_model: stabilityai/stable-diffusion-xl-base-1.0
instance_prompt: cinestyle, cinestyle cinematic still

---
#  CineStyle LoRA XL Base 

<Gallery />

## Model description 

This is a compact LoRA that allows the generation of moody cinematic stills inspired by Blade Runner 2049, Tenet and Co. in Stable Diffusion XL based models.

It was trained based on 65 images with a total of 1,450 steps. The base model is Stable Diffusion XL.

To use it, simply enter the trigger word “cinestyle” and&#x2F;or &quot;cinestyle cinematic still&quot; and then describe what you wanna see in your picture. 


## Trigger words

You should use `cinestyle` to trigger the image generation.

You should use `cinestyle cinematic still` to trigger the image generation.


## Download model

Weights for this model are available in Safetensors format.

[Download](/IcelosAI/Cinestyle_LoRA_XL_Base/tree/main) them in the Files & versions tab.