Upscale & Resize

The Upscale enhancement can do multiple things:

  1. CHANGE IMAGE SIZE: Change the size of your image to be up to 600% times bigger (1) or downscale the image to be a smaller size (0.5x for example) while preserving the intricate details from the original image.
  2. CHANGE FILE RESOLUTION: In the Dimensions options, if you change the size to be shown in inches or centimeters, you will see the Resolution field appear (will be in px/inch or px/cm). If left unchanged, Topaz Photo will retain the resolution of the original file.
  3. FIX ARTIFACTS: Upscale AI Models can also be used to help fix compression and/or resolution artifacts present in small low resolution files or files taken from a cellphone for example. Test the different AI Models to find the one best suited for your image's type of artifacts.

(1) Upscaling 6 times the original size will only be available if the original image is small. The "Max" scale for the image will be dependent on the starting image size. The "Max" will be 32 000 pixels on the longest side (about 8 feet wide, depending on the file resolution - dpi) or 2GB of file size, whichever is attained first.

Tips and Tricks

Processing a very large upscale will take up a lot of graphic memory and processing time to export. Here are tips to optimize your upscale:

Scale/Dimensions/Resolution

Use the Upscale Controls Panel to adjust the Scale, Dimensions and Resolution.

  • SCALE: Choose the scale factor here. Make sure to only upscale to a file size you need. If you don't need the image bigger in size and only want the compression/resolution artifacts fixed, choose 1x scale.
  • DIMENSIONS: If you need a specific size, you can change the size in the Dimensions field. You can toggle from Pixel, to inches, to centimeters.
  • RESOLUTION: Choose from pixels, inches, or centimeters to specify dimensions. Set values for each side or enter the overall DPI (dots per inch) or PPI (pixels per inch). Use this tool to define dimensions for printing purposes. The pixels/inch and pixels/centimeter textboxes only appear when the dimensions are set to those units. Example, to see the DPI and set it to 300 dpi (dots per inch), make sure to select the Dimensions to be in inches.

AI Models

The AI models rely heavily on existing image quality. They may generate unwanted artifacts if the resolution of the starting image is too low. Upscale from an original image whenever possible rather than a copy that has been compressed.

Make sure to use the appropriate model for your image. The models are not interchangeable and using the wrong model may deliver unwanted results.

Standard

The Standard model is the go-to choice for general use unless the image clearly belongs to one of the other categories. It is tailored for medium-quality images from point-and-shoot cameras and older smartphones.

Use the Standard model for highly detailed AI-generated images from Midjourney, Dall-E, and similar tools.

Standard

Standard V2

Use the Standard V2 model to have a more generative output than the original standard model.

High Fidelity

Use this model for images that are already high-quality to preserve intricate details. Images from DSLRs and most modern smartphones will fall under this category. Use it for highly detailed AI-generated art as well.

High Fidelity

High Fidelity V2

Use the High Fidelity V2 model to intentionally preserve original grain or noise for aesthetic purposes.

Low Resolution

The Low Resolution model prioritizes preserving overall quality rather than fine details. Use it to upscale low-quality images that have been highly compressed.

Low Resolution

Graphics

The Graphics model upscales simple illustrations such as cartoons.

Graphics


Standard MAX

This model is built on new architecture, producing very high-quality outputs similar to generative upscaling models in Gigapixel AI while significantly lowering the hardware requirements and high processing time. Standard MAX can run on a wide range of devices — making advanced restoration more accessible than ever.

Standard MAX delivers outstanding results across all types of input images, but it truly shines on low-resolution photos where it generates incredible photorealistic detail that surpasses anything currently available in Topaz Photo. The results are clean and natural, with refined textures that are hard to achieve in other AI upscaling models. While we highly recommend trying it on low-res images for truly jaw-dropping outputs, Standard Max also performs exceptionally well on medium and high-quality images, making it a versatile and powerful choice for a wide range of upscaling needs.


Before and after on a low resolution image

image
image

Recover V3

Released in Topaz Photo v1.2.0, Recover v3 represents our highest quality approach to AI upscaling. It’s built to produce realistic photographic texture, not artificial sharpening or overly processed results.

Instead of simply increasing resolution, Recover 3 focuses on reconstructing fine surface detail for things like fur, skin texture, foliage patterns, and fabric structure. This makes it especially strong for nature photography, wildlife, and portraits, where realism matters most.

Availability


Recover v3 is the newest model available within the Upscale models. Recover 3 works with batch processing, and can run both locally and in the cloud. For most users, cloud rendering is recommended for speed. Local processing is ideal when privacy, offline workflows, or unattended batch jobs are more important than raw performance.


Here are the Recover v3 System Requirements:


For Windows

For macOS

  • Optimal input size: 1024 × 1024 or smaller
  • Works best on lower-quality images that still contain usable detail.
  • Larger images are supported, but renders more slowly
  • Not recommended for highly structured, high-grain, or high-resolution images. Isn’t ideal for images dominated by text or logos, which diffusion models still struggle to reproduce accurately.

Creativity Control

The Creativity Control lets you decide how aggressively the model adds texture. Lower settings keep results conservative and natural, while higher settings push more fine detail into the image.

This flexibility gives you control depending on the source quality and the look you’re trying to achieve. Keep in mind that higher creativity levels can introduce background artifacts, so it’s best used intentionally.

Additional Settings

Minor Denoise

Low light conditions, sensor limitations, or compression artifacts can cause slight noise. The Minor Denoise setting removes this noise before the image is upscaled.

Minor Deblur

Use the Minor Deblur setting to avoid emphasizing existing blurriness caused by camera shake, motion blur, or a lack of focus when the image is captured.

Fix Compression

Visual distortions and artifacts may appear when images are compressed to make them more space-efficient.

Preserve Text

Wonder