> ## Documentation Index
> Fetch the complete documentation index at: https://dripart-codex-simplify-get-started.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Create your first image

> Run an image workflow in ComfyUI, change its prompt, and save the result.

A **workflow** is a set of connected steps that creates an image, video, or other result. You can start with a ready-made template. You do not need to build a workflow or understand every node before your first run.

## Before you begin

You can follow this guide in [Comfy Cloud](/get_started/cloud) first. Open it in your browser to use an image template without installing ComfyUI or downloading a local model. If you prefer to run ComfyUI on your own computer, [install Comfy Desktop](/installation/desktop/overview) and follow the same steps. A local template may ask you to download models.

<Note>
  This guide uses an image template as an example. Template names and available models can change as ComfyUI is updated.
</Note>

## Run an image template

<Steps>
  <Step title="Open a template">
    In ComfyUI, click **Templates** in the left sidebar, or open **Workflow** → **Browse Workflow Templates**. Choose an image generation template that accepts a text prompt.

    A template loads a complete workflow onto the canvas. Each box is a **node** that handles one part of the process. The lines between boxes pass information from one node to the next.
  </Step>

  <Step title="Resolve any missing models">
    In Comfy Cloud, choose a template whose models are ready to use. On a local installation, ComfyUI may report missing models. Follow its download prompt before running the workflow.

    Comfy Desktop downloads supported files from the prompt. With portable or manual installations, the browser may download the files instead. Save each file in the model folder shown in the prompt, then refresh the model list or restart ComfyUI. See [Workflow Templates](/interface/features/template#model-storage-location) for details.
  </Step>

  <Step title="Change the prompt and run">
    Find the text input in the loaded workflow and replace it with a short description, such as `a red bicycle parked beside a café, watercolor illustration`. Depending on the template, this may be a text field inside a node or an input panel.

    Click **Run**. ComfyUI will process the workflow and show the result in its output node or the queue panel. The first run can take longer while the model loads.
  </Step>

  <Step title="Save and try one change">
    Right-click the finished image and select **Save image**. Change a few words in the prompt and click **Run** again to see what changes.
  </Step>
</Steps>

## What you just used

* **Template**: A ready-made workflow you can run and edit.
* **Node**: A box that performs a step, such as loading a model, reading a prompt, or saving an image.
* **Model**: The downloaded AI file used to generate the result. Local workflows need the models they reference.
* **Link**: A connection that passes information between nodes.

You can learn more in [Workflows](/basic-concepts/workflow), or follow the [text-to-image tutorial](/tutorials/basic/text-to-image) to explore the image generation steps in detail.

If you are unsure whether your template uses a downloaded model or a hosted node, see [Where a workflow runs](/get_started/where-work-runs).

## If the workflow does not run

* **Missing model**: Open the missing-model details and install the requested file. If a model dropdown is empty after downloading, check its folder and refresh or restart ComfyUI.
* **Missing node**: Try a built-in template first. Imported or custom-node templates may require extra nodes. See [install custom nodes](/installation/install_custom_node).
* **No visible result**: Check the queue panel for progress or errors, and look for an output node such as **Save Image**.

For other errors, use [Fix a workflow that will not run](/get_started/fix-a-workflow).