> For the complete documentation index, see [llms.txt](https://runyourai.gitbook.io/eng-runyourai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://runyourai.gitbook.io/eng-runyourai/tutorial/using-stable-diffusion.md).

# Using Stable-diffusion

<figure><img src="/files/ZCMrWTSXXTzIkmDoWEd8" alt=""><figcaption></figcaption></figure>

{% hint style="info" %} <mark style="color:red;">What is</mark> <mark style="color:red;"></mark><mark style="color:red;">**Stable Diffusion**</mark><mark style="color:red;">?</mark>

It's an image generation model based on deep learning technology.&#x20;

You can generate various images based on the given prompts and settings.
{% endhint %}

**What can you do with Stable Diffusion?**

✔️Generate high-quality images that are difficult to distinguish from reality.

✔️Create new images based on real images.

✔️Use your imagination to generate new images.

**Try Stable Diffusion now using the&#x20;**<mark style="background-color:yellow;">**template feature**</mark>**&#x20;provided by Runyour AI!**

***

## Step 1. Machine Rental / Template Selection&#x20;

<figure><img src="/files/uVhMg4CeHiseRfwTI2OP" alt=""><figcaption></figcaption></figure>

* When you select your desired machine from the GPU Cloud list, you will proceed to the template selection step. \
  Choose the <mark style="background-color:yellow;">**Stable Diffusion**</mark> template and click the rent button at the top.
* You can view the rented machine on <mark style="background-color:yellow;">My Page > Dashboard</mark>.

## Step 2. Access to WebUI

<figure><img src="/files/EFIXRjlMkKU6lrR5mgND" alt=""><figcaption></figcaption></figure>

* If you click on the rented machine, you can check the SSH / Web UI access function and the PEM key download function.

<figure><img src="/files/Ju0t6yMyAT8BHXHGTHeO" alt=""><figcaption></figcaption></figure>

* Click the SSH / WEB button among the two buttons.

<figure><img src="/files/GTNW8R24tbHABwIgFjR2" alt=""><figcaption></figcaption></figure>

* Among the two access methods, click the 'HTTP' > 'Stable Diffusion Web' button to access the Web UI directly.

<figure><img src="/files/WdZM0MbS22ZtWZpn0YGB" alt=""><figcaption><p>Stable diffusion Web UI Screen</p></figcaption></figure>

{% hint style="info" %}
Please refer to the '[Remote Access via SSH](/eng-runyourai/tutorial/remote-access-via-ssh-common.md)' about the SSH access
{% endhint %}

* Access method differs by the user's OS, you can access referring to the following:

> 1\) In case of <mark style="color:red;">**Mac OS,**</mark> run the **Terminal**
>
> 2\) In case of <mark style="color:red;">**Window OS**</mark>, run the **CMD, Powershell or other terminal program.**

## Step 3. Run Stable Diffusion WebUI

### Utilize 'txt2img' feature

Let's generate an image using the prominent feature 'txt2img'.

<figure><img src="/files/WpkG5LW94xF5zsOriesi" alt=""><figcaption></figcaption></figure>

<mark style="color:red;">**1) Enter instructions for generating the image in the Prompt.**</mark>&#x20;

For stable-diffusion-webui, it runs on port 7860 on the respective machine.

```
Prompt Example

A magical forest with glowing trees, ethereal creatures, sparkling fairy lights, and a misty atmosphere.
```

<mark style="color:red;">**2) In the Negative prompt, list elements that should be excluded from the image.**</mark>

```
Negative Prompt Example

Modern buildings, cars, dark shadows, blurry details, dull colors.
```

<mark style="color:red;">**3) Review detailed settings under Generation and click the Generate button at the top right to create the image.**</mark>

<figure><img src="/files/1gyT2dwknZtKVNXQMxbg" alt=""><figcaption></figcaption></figure>