> For the complete documentation index, see [llms.txt](https://developer.topazlabs.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://developer.topazlabs.com/getting-started/model-selection.md).

# Model Selection

Select the best model for your use case

Models are organized by what they do. In the sidebar, locate **Image Models** or **Video Models** and choose the family that matches your goal, such as Upscaling, Denoising, or Sharpening. Each family page lists every model in it, with a short description of when to use each one.

{% hint style="success" %}
**Not sure where to start?** Use **Wonder 4** for images and **Starlight Precise 2.6** for video. These are our top generative upscaling models for image and video. Both handle a wide range of sources well, and you can switch to a more specialized model if the result calls for it. See [Popular Models](https://developer.topazlabs.com/popular-models.md) for our top picks.
{% endhint %}

### Image Models

<table data-view="cards"><thead><tr><th></th><th></th><th></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><strong>Upscaling &#x26; Enhancement</strong></td><td>Increase resolution and recover detail.</td><td>Start with <strong>Wonder 4</strong>, or <strong>Standard 2</strong> for more faithful-to-the-original upscaling.</td><td><a href="https://developer.topazlabs.com/image-models/upscaling-and-enhancement.md">Upscaling &amp; Enhancement</a></td></tr><tr><td><strong>Denoising</strong></td><td>Remove noise and grain.</td><td>Start with <strong>Denoise Max</strong>.</td><td><a href="https://developer.topazlabs.com/image-models/denoising.md">Denoising</a></td></tr><tr><td><strong>Sharpening</strong></td><td>Fix blur and soft focus.</td><td>Start with <strong>Super Focus 3.</strong></td><td><a href="https://developer.topazlabs.com/image-models/sharpening.md">Sharpening</a></td></tr><tr><td><strong>Removal &#x26; Cleanup</strong></td><td>Remove backgrounds, dust, and scratches.</td><td>See the family page.</td><td><a href="https://developer.topazlabs.com/image-models/removal-and-cleanup.md">Removal &amp; Cleanup</a></td></tr><tr><td><strong>Color, Lighting &#x26; HDR</strong></td><td>Correct exposure, white balance, and color.</td><td>Start with <strong>Adjust Lighting 2</strong>.</td><td><a href="https://developer.topazlabs.com/image-models/color-and-lighting.md">Color, Lighting, &amp; HDR</a></td></tr></tbody></table>

### Video Models

<table data-view="cards"><thead><tr><th></th><th></th><th></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><strong>Upscaling &#x26; Enhancement</strong></td><td>Increase resolution and restore footage.</td><td>Start with <strong>Starlight Precise 2.6</strong>, or <strong>Proteus</strong> for more faithful-to-the-original upscaling.</td><td><a href="https://developer.topazlabs.com/video-models/upscaling-and-enhancement.md">Upscaling &amp; Enhancement</a></td></tr><tr><td><strong>Denoising</strong></td><td>Remove noise and grain.</td><td>Start with <strong>Nyx</strong>.</td><td><a href="https://developer.topazlabs.com/video-models/denoising.md">Denoising</a></td></tr><tr><td><strong>Frame Interpolation</strong></td><td>Create slow motion and change frame rates.</td><td>Start with <strong>Apollo</strong>.</td><td><a href="https://developer.topazlabs.com/video-models/frame-interpolation.md">Frame Interpolation</a></td></tr><tr><td><strong>Video Utilities &#x26; HDR</strong></td><td>Motion deblur and other specialized fixes.</td><td>See the family page.</td><td><a href="https://developer.topazlabs.com/video-models/other-utilities.md">Video Utilities &amp; HDR</a></td></tr></tbody></table>

### Combining Models

You can run more than one model on the same file. For images, send the output of one request as the input to the next, for example **Denoise Max** followed by **Wonder 4** for a very noisy photo that also needs upscaling. For video, add several models to the `filters` array of a single request, for example **Proteus** followed by **Apollo** for a video that needs upscaling, where you'd also like to adjust the fps value.

#### Image Example: Denoise Max + Wonder 4

Run **Denoise Max** first, then pass the `download_url` from `GET /image/v1/download/{process_id}` as `source_url` in the **Wonder 4** request.

{% tabs %}
{% tab title="cURL" %}

```bash
# 1. Denoise the image with Denoise Max
curl -X POST "https://api.topazlabs.com/image/v1/denoise-gen/async" \
  -H "X-API-Key: $TOPAZ_API_KEY" \
  -F "image=@noisy-photo.jpg" \
  -F "model=Denoise Max" \
  -F "strength=0.8" \
  -F "output_format=png"

# Poll GET /image/v1/status/{process_id} until Completed,
# then fetch download_url from GET /image/v1/download/{process_id}.

# 2. Upscale the denoised output with Wonder 4 using source_url
curl -X POST "https://api.topazlabs.com/image/v1/enhance-gen/async" \
  -H "X-API-Key: $TOPAZ_API_KEY" \
  -F "source_url=$DENOISED_DOWNLOAD_URL" \
  -F "model=Wonder 4" \
  -F "output_width=1280"
```

{% endtab %}

{% tab title="JavaScript" %}

```javascript
import fs from 'fs';
import FormData from 'form-data';

const HEADERS = { 'X-API-Key': process.env.TOPAZ_API_KEY };
const IMAGE_BASE = 'https://api.topazlabs.com/image/v1';

// 1. Denoise the image with Denoise Max
const denoiseForm = new FormData();
denoiseForm.append('image', fs.createReadStream('noisy-photo.jpg'));
denoiseForm.append('model', 'Denoise Max');
denoiseForm.append('strength', '0.8');
denoiseForm.append('output_format', 'png');

const denoiseRes = await fetch(`${IMAGE_BASE}/denoise-gen/async`, {
  method: 'POST',
  headers: { ...HEADERS, ...denoiseForm.getHeaders() },
  body: denoiseForm,
});
console.log(await denoiseRes.json());

// Poll GET /image/v1/status/{process_id} until Completed,
// then fetch download_url from GET /image/v1/download/{process_id}.

// 2. Upscale the denoised output with Wonder 4 using source_url
const upscaleForm = new FormData();
upscaleForm.append('source_url', denoisedDownloadUrl);
upscaleForm.append('model', 'Wonder 4');
upscaleForm.append('output_width', '1280');

const upscaleRes = await fetch(`${IMAGE_BASE}/enhance-gen/async`, {
  method: 'POST',
  headers: { ...HEADERS, ...upscaleForm.getHeaders() },
  body: upscaleForm,
});
console.log(await upscaleRes.json());
```

{% endtab %}

{% tab title="Python" %}

```python
import os
import requests

HEADERS = {"X-API-Key": os.environ["TOPAZ_API_KEY"]}
IMAGE_BASE = "https://api.topazlabs.com/image/v1"

# 1. Denoise the image with Denoise Max
with open("noisy-photo.jpg", "rb") as image:
    denoise_res = requests.post(
        f"{IMAGE_BASE}/denoise-gen/async",
        headers=HEADERS,
        files={"image": image},
        data={
            "model": "Denoise Max",
            "strength": "0.8",
            "output_format": "png",
        },
    )
denoise_res.raise_for_status()
print(denoise_res.json())

# Poll GET /image/v1/status/{process_id} until Completed,
# then fetch download_url from GET /image/v1/download/{process_id}.

# 2. Upscale the denoised output with Wonder 4 using source_url
upscale_res = requests.post(
    f"{IMAGE_BASE}/enhance-gen/async",
    headers=HEADERS,
    files={
        "source_url": (None, denoised_download_url),
        "model": (None, "Wonder 4"),
        "output_width": (None, "1280"),
    },
)
upscale_res.raise_for_status()
print(upscale_res.json())
```

{% endtab %}
{% endtabs %}

#### Video Example: Proteus + Apollo

Include both filter objects in the `filters` array of a single request. They run in order: **Proteus** (`prob-4`) upscales the video to 4K, and **Apollo** (`apo-8`) interpolates the frame rate to 60 fps.

{% tabs %}
{% tab title="cURL" %}

```bash
curl -X POST "https://api.topazlabs.com/video/express" \
  -H "X-API-Key: $TOPAZ_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "source": { "container": "mp4" },
    "filters": [
      { "model": "prob-4", "auto": "Auto" },
      { "model": "apo-8", "fps": 60 }
    ],
    "output": {
      "resolution": { "width": 3840, "height": 2160 },
      "frameRate": 60
    }
  }'
```

{% endtab %}

{% tab title="JavaScript" %}

```javascript
const payload = {
  source: {
    container: 'mp4',
  },
  filters: [
    { model: 'prob-4', auto: 'Auto' },
    { model: 'apo-8', fps: 60 },
  ],
  output: {
    resolution: { width: 3840, height: 2160 },
    frameRate: 60,
  },
};

const response = await fetch('https://api.topazlabs.com/video/express', {
  method: 'POST',
  headers: {
    'X-API-Key': process.env.TOPAZ_API_KEY,
    'Content-Type': 'application/json',
  },
  body: JSON.stringify(payload),
});

if (!response.ok) {
  throw new Error(`Request failed: ${response.status} ${await response.text()}`);
}

console.log(await response.json());
```

{% endtab %}

{% tab title="Python" %}

```python
import os
import requests

payload = {
    "source": {"container": "mp4"},
    "filters": [
        {"model": "prob-4", "auto": "Auto"},
        {"model": "apo-8", "fps": 60},
    ],
    "output": {
        "resolution": {"width": 3840, "height": 2160},
        "frameRate": 60,
    },
}

response = requests.post(
    "https://api.topazlabs.com/video/express",
    headers={"X-API-Key": os.environ["TOPAZ_API_KEY"]},
    json=payload,
)
response.raise_for_status()
print(response.json())
```

{% endtab %}
{% endtabs %}

#### Video Example: Starlight Precise 2.6 + Hyperion 2.5

Include both filter objects in the `filters` array of a single request. They run in order: **Starlight Precise 2.6** (`slp-2.6`) upscales the video to 4K, and **Hyperion 2.5** (`hyp-2.5`) converts the upscaled result from SDR to HDR. Hyperion 2.5 outputs 10-bit ProRes, so set `output.container` to `mov`.

{% tabs %}
{% tab title="cURL" %}

```bash
curl -X POST "https://api.topazlabs.com/video/express" \
  -H "X-API-Key: $TOPAZ_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "source": { "container": "mp4" },
    "filters": [
      { "model": "slp-2.6", "sharpness": 3.0 },
      { "model": "hyp-2.5" }
    ],
    "output": {
      "resolution": { "width": 3840, "height": 2160 },
      "container": "mov"
    }
  }'
```

{% endtab %}

{% tab title="JavaScript" %}

```javascript
const payload = {
  source: {
    container: 'mp4',
  },
  filters: [
    { model: 'slp-2.6', sharpness: 3.0 },
    { model: 'hyp-2.5' },
  ],
  output: {
    resolution: { width: 3840, height: 2160 },
    container: 'mov',
  },
};

const response = await fetch('https://api.topazlabs.com/video/express', {
  method: 'POST',
  headers: {
    'X-API-Key': process.env.TOPAZ_API_KEY,
    'Content-Type': 'application/json',
  },
  body: JSON.stringify(payload),
});

if (!response.ok) {
  throw new Error(`Request failed: ${response.status} ${await response.text()}`);
}

console.log(await response.json());
```

{% endtab %}

{% tab title="Python" %}

```python
import os
import requests

payload = {
    "source": {"container": "mp4"},
    "filters": [
        {"model": "slp-2.6", "sharpness": 3.0},
        {"model": "hyp-2.5"},
    ],
    "output": {
        "resolution": {"width": 3840, "height": 2160},
        "container": "mov",
    },
}

response = requests.post(
    "https://api.topazlabs.com/video/express",
    headers={"X-API-Key": os.environ["TOPAZ_API_KEY"]},
    json=payload,
)
response.raise_for_status()
print(response.json())
```

{% endtab %}
{% endtabs %}

To compare models on your own files before you integrate, use the [Model Playground](https://playground.topazlabs.com).

### Next Steps

<table data-view="cards"><thead><tr><th></th><th></th><th></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><i class="fa-tag">:tag:</i></td><td><h4>Model Pricing</h4></td><td>Learn how we price our models.</td><td><a href="https://developer.topazlabs.com/getting-started/model-pricing.md">Model Pricing</a></td></tr></tbody></table>


---

# 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://developer.topazlabs.com/getting-started/model-selection.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.
