---
tags:
- text-to-image
- stable-diffusion
- lora
- diffusers
- template:sd-lora
- StableDiffusionXLPipeline
- Inference Endpoints
- sdxl
- realism
- tensorart
widget:
- text: >-
    A man dressed in sunglasses and brown jacket, in the style of cypherpunk,
    timeless beauty, exacting precision, uhd image, aleksandr deyneka, matte
    background, leather/hide  --ar 67:101 --v 5
  output:
    url: images/1.png
- text: >-
    A studio portrait of a brunette model wearing a dress in front of a natural
    background --v 6.0 --style raw
  output:
    url: images/2.png
- text: >-
    Man in the style of dark beige and brown, uhd image, youthful protagonists,
    nonrepresentational 
  output:
    url: images/3.png
base_model: stabilityai/stable-diffusion-xl-base-1.0
instance_prompt: PhotoRealism
license: creativeml-openrail-m
---
# Canopus Photo Realism

<Gallery />

**The model is still in the training phase. This is not the final version and may contain artifacts and perform poorly in some cases.**

## Working Demo Canopus-Realism

| Parameter           | Description                                          |
|---------------------|------------------------------------------------------|
| **Space Name**      | Canopus-Realism                                      |
| **Hosted on**       | [⚡Hugging Face Spaces⚡](https://huggingface.co/spaces/prithivMLmods/Canopus-Realism) |
| **Default Theme**   |      Light by Defalut / Can be altered                                         |
| **Description**     | Canopus-Realism provides realistic image generation capabilities with a focus on high-quality outputs. |

## Model description 

Canopus Photo Realism

Image Processing Parameters 

| Parameter                 | Value  | Parameter                 | Value  |
|---------------------------|--------|---------------------------|--------|
| LR Scheduler              | constant | Noise Offset              | 0.03   |
| Optimizer                 | AdamW  | Multires Noise Discount   | 0.1    |
| Network Dim               | 64     | Multires Noise Iterations | 10     |
| Network Alpha             | 32     | Repeat                    | 20     |
| Epoch                     | 20     | Save Every N Epochs       | 1      |

## SETTING-UP

```py
    pipe = StableDiffusionXLPipeline.from_pretrained(
        "-------------xxxxxxxxx----------",
        torch_dtype=torch.float16,
        use_safetensors=True,
    )
                    (or)
-----------------------------------------------------------
    pipe = StableDiffusionXLPipeline.from_pretrained(
        "stabilityai/stable-diffusion-xl-base-1.0",
        torch_dtype=torch.float16,
        use_safetensors=True,
    )
    pipe.scheduler = EulerAncestralDiscreteScheduler.from_config(pipe.scheduler.config)
    
    pipe.load_lora_weights("prithivMLmods/Canopus-Realism-LoRA", weight_name="Canopus-Realism-LoRA.safetensors", adapter_name="rlms")
    pipe.set_adapters("rlms")
    pipe.to("cuda")
```
## Trigger prompts

    A man dressed in sunglasses and brown jacket, in the style of cypherpunk, timeless beauty, exacting precision, uhd image, aleksandr deyneka, matte background, leather/hide  --ar 67:101 --v 5


    A studio portrait of a brunette model wearing a dress in front of a natural background --v 6.0 --style raw


    Man in the style of dark beige and brown, uhd image, youthful protagonists, nonrepresentational 

| Parameter       | Value                                                                                 |
|-----------------|---------------------------------------------------------------------------------------|
| Prompt          | Closeup of blonde woman depth of field, bokeh, shallow focus, minimalism, fujifilm xh2s with Canon EF lens, cinematic --ar 85:128 --v 6.0 --style raw |
| Sampler         | euler                                                                                 |

## Download model

Weights for this model are available in Safetensors format.

[Download](/prithivMLmods/Canopus-Realism-LoRA/tree/main) them in the Files & versions tab.