# Upscaling Images

Fater includes specialized AI models designed to **upscale your images**, meaning they increase the image's resolution (pixel dimensions) while intelligently enhancing details and attempting to maintain or improve overall quality. This is invaluable for preparing images for high-resolution displays, print, or simply to recover detail from smaller source images.

Upscaling is typically done in the **Image Editor**.

***

### When to Use AI Upscaling

* Increasing the resolution of existing renders, photographs, or AI-generated images.
* Adding sharpness and clarity to slightly blurry or soft images.
* Revealing finer details that might be lost in lower-resolution versions.
* Preparing images for large-format printing or display.

***

### Workflow for Upscaling Images

1. **Prepare Your Image Layer & Generation Area:**
   * Ensure the image content you want to upscale is visible within the **Generation Area Bounding Box** on your **Inpaint Canvas**. The AI will process the content within this box.
   * **Mask (Optional - for Selective Upscaling):** If you only want to upscale a part of an image, create a **Mask** over that area.
2. **Select an Upscaling AI Model:**
   * Go to the **Model Selector** (Left Sidebar), choose the **"Upscale"** category (🖼️ icon), and select a model.
3. **Configure Upscaling Parameters:**
   * Adjust settings in the Left Sidebar. Key parameters include:
     * **Prompt & Negative Prompt:** Can sometimes refine upscale style.
     * **Scale Factor & Output Megapixels (Resolution Control):**
       * **Scale:** A direct multiplier for the input dimensions (e.g., 2x, 4x).
       * **Output Megapixels:** Allows you to target a specific final resolution in millions of pixels.
       * **How they interact:** Fater determines the AI model's internal processing size based on a combination of:
         1. The current dimensions of your **Generation Area Bounding Box** (this is the "input image" for the upscaler).
         2. The chosen **Scale** factor.
         3. The target **Output MegaPixels**.\
            The system intelligently calculates the optimal internal dimensions for the model to achieve an output that respects both your desired scale and target megapixel count as closely as possible. This means you don't usually need to precisely calculate input sizes yourself; set your desired final output characteristics.
     * **Creativity / Resemblance:** Balance new detail vs. faithfulness to the original.
     * **Sharpen, Dynamic, Tile Size, More Details, Hand Fix, Checkpoint:** Further refine the process.
     * **Mask Alpha Radius (for masked upscaling):** Controls edge blending if a mask is used.
     * *(See* [Understanding Generation Parameters](/fater-ai-docs/understanding-generation-parameters.md) *and the* [AI Models](/fater-ai-docs/reference-materials/ai-models.md) *for details).*
4. **Initiate Upscaling:**
   * Click the **Generate** button (✨) in the **Floating Prompt Area**.
5. **Review and Use Results:**
   * The first upscaled image is automatically added as a new layer.
   * Other variations (if any) appear in the **Generation Task List**.

***

**Example Scenario : Preparing a Product Shot for a Large Banner**

1. **Image Layer:** You have a good but medium-resolution product photo on a layer.
2. **Generation Area:** Set the bounding box to cover the entire product photo.
3. **Model:** Select "High \[Upscale]".
4. **Parameters:**
   * Set Scale to 4x and Output Megapixels to your target banner resolution.
5. **Generate.**
6. Review the new high-resolution layer.

***

AI Upscaling in Fater can significantly enhance the quality and usability of your images, making them suitable for a wider range of applications.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://fater.gitbook.io/fater-ai-docs/generating-images-with-ai/upscaling-images.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.