> For the complete documentation index, see [llms.txt](https://cranford-tech.gitbook.io/replicate-ai-and-ml-models-plugin/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://cranford-tech.gitbook.io/replicate-ai-and-ml-models-plugin/retrieving-predictions/get-prediction-workflow-actions.md).

# Get Prediction Workflow Actions

Each of the models that come with the plugin have a "Get Prediction" workflow action:

* Get Prediction - Stable Diffusion
* Get Prediction - GFPGAN
* Get Prediction - SDXL

The below video tutorial shows how to set up your Bubble database and use the "Get Prediction - GFPGAN" action from 07:00. 

{% embed url="<https://www.youtube.com/watch?v=-VlztwBGzns>" %}

You can run the "Get Prediction" workflow action either on page load or as a backend workflow.&#x20;

If you run it on page load, you'll need to:

* Set the type of data on the page to "prediction"&#x20;
* Send the relevant prediction to the page you're loading
* EG, you can see in the below that we're creating a prediction in a workflow action and then sending that prediction to the "prediction-result" page

**IMPORTANT:** You will need to add a pause before navigating to the "prediction" result page to give Replicate time to create the prediction. It may take a bit of trial and error to figure out exactly how long the pause should be.&#x20;

<figure><img src="/files/kA2SOkJ583CIUw3n8y2q" alt=""><figcaption></figcaption></figure>

You can then add a "Get Prediction" workflow action on page load to the "predictions-result" page and save down the relevant output:

<div align="left"><figure><img src="/files/JfNv27all9eSa1GwZl4G" alt="" width="563"><figcaption></figcaption></figure></div>

<div align="left"><figure><img src="/files/b3JOdt4uwqWS496CwNb9" alt="" width="563"><figcaption></figcaption></figure></div>