> For the complete documentation index, see [llms.txt](https://docs.viesus.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.viesus.com/features/features/ai-super-resolution.md).

# AI Upscaling

AI upscaling by up to 4× per pass (up to 16× total with two-pass mode). Multiple speed/quality variants — the scene detection system can select the optimal model for the detected content type.

{% hint style="warning" %}
**Requires:** NVIDIA GPU (Ampere architecture or newer, ≥ 8 GB VRAM) for production throughput.
{% endhint %}

## Example

{% columns %}
{% column %}
**Source resolution**

<figure><img src="https://919968377-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FG3eQlcnlDKtn9yl3mxsx%2Fuploads%2Fgit-blob-2aac38d3a6989f640b75c2aeecf11934ec13a3c3%2Fplaceholder.svg?alt=media" alt="Source — AI upscaling"><figcaption></figcaption></figure>
{% endcolumn %}

{% column %}
**4× AI upscaling**

<figure><img src="https://919968377-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FG3eQlcnlDKtn9yl3mxsx%2Fuploads%2Fgit-blob-2aac38d3a6989f640b75c2aeecf11934ec13a3c3%2Fplaceholder.svg?alt=media" alt="Upscaled 4× — AI upscaling"><figcaption></figcaption></figure>
{% endcolumn %}
{% endcolumns %}

## Parameters

<table><thead><tr><th width="165.5999755859375">Parameter</th><th width="97.5999755859375">Section</th><th width="85.4000244140625">Value</th><th width="95">Default</th><th>Description</th></tr></thead><tbody><tr><td><code>ResizeMode</code></td><td><code>Resize</code></td><td>0 – 12</td><td>5</td><td>Mode selection (see below)</td></tr><tr><td><code>ResizeFacDetMode</code></td><td><code>Resize</code></td><td>0 – 3</td><td>1</td><td>Size calculation mode. <strong>0</strong> = factor, <strong>1</strong> = small side, <strong>2</strong> = width, <strong>3</strong> = height</td></tr><tr><td><code>ResizeFactor</code></td><td><code>Resize</code></td><td>0.1 – 8.0</td><td>4.0</td><td>Resize factor (used when <code>ResizeFacDetMode</code> = 0)</td></tr><tr><td><code>SupResThresh</code></td><td><code>Resize</code></td><td>Float</td><td>2.0</td><td>SR only activates above this scale factor</td></tr><tr><td><code>SupRes2xThresh</code></td><td><code>Resize</code></td><td>Float</td><td>3.0</td><td>Scale-factor threshold up to which the 2× model is used; above it, the 4× model is applied</td></tr></tbody></table>

## Modes

<table><thead><tr><th width="260.4000244140625">Value</th><th>Description</th></tr></thead><tbody><tr><td><strong>0</strong> — Default</td><td>Auto — selects algorithm based on content</td></tr><tr><td><strong>1</strong> — Bicubic</td><td>Bicubic interpolation</td></tr><tr><td><strong>2</strong> — Biquadratic</td><td>Biquadratic interpolation</td></tr><tr><td><strong>3</strong> — Bilinear</td><td>Bilinear interpolation</td></tr><tr><td><strong>4</strong> — Nearest Neighbour</td><td>Nearest neighbour interpolation</td></tr><tr><td><strong>5</strong> — AI SR ×4, quality</td><td>AI-based 4× super-resolution</td></tr><tr><td><strong>6</strong> — Default — no SR</td><td>Default without super-resolution</td></tr><tr><td><strong>7</strong> — AI SR ×2 / ×4, quality</td><td>Auto-selects 2× or 4× based on <code>ResizeFactor</code></td></tr><tr><td><strong>8</strong> — AI SR ×2 / ×4, fast</td><td>Fast variant of mode 7</td></tr><tr><td><strong>9</strong> — AI SR ×4, fast</td><td>Fast 4× super-resolution</td></tr><tr><td><strong>10</strong> — AI SR ×2 fast / ×4 fast</td><td>Fast 2× or 4× (alternate path)</td></tr><tr><td><strong>11</strong> — AI SR ×4 with AR</td><td>AI 4× super-resolution with integrated artifact removal</td></tr><tr><td><strong>12</strong> — AI SR ×2 / ×4 Legacy</td><td>Legacy 2× and 4× super-resolution model</td></tr></tbody></table>

For the full Resize section reference (including `ResizeFacDetMode`, `ResizePixSize`, `ResizeOn`), see the [Parameter Reference → Resizing](/configuration/parameter-reference/resizing.md).

{% hint style="success" %}
**Recommended:** Mode 7 for quality-focused production. Mode 11 if input images have visible JPEG artifacts.
{% endhint %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

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

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.viesus.com/features/features/ai-super-resolution.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.