Topaz Gigapixel AI Pro

Overview

Gigapixel exists because conventional enlargement cannot invent what is not in the file. Traditional interpolation averages between the pixels it already has, which is why an upscaled image looks soft and plasticky. This tool instead runs trained models that reconstruct plausible edges, textures and detail based on what millions of examples imply should be there, and the difference at four or six times enlargement is not subtle.

Model selection carries most of the result. Separate models are trained for clean high quality photographs, for heavily compressed web images, for low resolution sources, for scanned prints and film, for computer generated art and illustration, and for photographs where recovering facial detail matters more than anything else. The comparison view runs several models on a section of your own image at full resolution so the choice is made by looking rather than by guessing.

Around the models sit the controls that make output usable rather than merely large. Face recovery strength keeps portraits from turning waxy, detail and sharpening sliders control how aggressive the reconstruction is, noise and compression suppression run before enlargement rather than after, and a fidelity control governs how closely the result stays to the original rather than inventing texture that was never there.

The application handles production work as well as single images. A batch queue processes an entire folder unattended with the same settings, output can be sized by scale factor, by absolute pixel dimensions or by print dimensions at a target resolution, and the results write out as TIFF, PNG, JPEG or DNG. It also installs as a plugin into the common photo editors, so an enlargement can happen inside an existing retouching pass rather than as a separate round trip.

Key Features

• Trained enlargement models reconstructing detail rather than interpolating pixels
• Separate models for photographs, compressed sources, low resolution, prints, art and faces
• Side by side model comparison at full resolution on a section of your own image
• Face recovery with adjustable strength for portrait work
• Noise and compression artefact suppression applied before enlargement
• Fidelity control governing how far the result may depart from the original
• Output by scale factor, pixel dimensions or print size at a target resolution
• Batch queue for processing a whole folder unattended
• Output to TIFF, PNG, JPEG and DNG with colour profile preservation
• Plugin integration into the common photo editors for in place enlargement

What Changed in This Build

• Improved texture reconstruction on fabric, foliage and skin
• Faster processing on current generation GPUs
• Face recovery reworked to avoid the waxy look at high strength
• Batch queue reliability improved on very large folders

System Requirements

Processor	Intel or AMD 64-bit, 4 cores or more recommended
Memory	16 GB recommended for large enlargements
Graphics	GPU with 6 GB VRAM recommended, processing runs on CPU without one
Storage	12 GB for install with all models
Display	1920 x 1080 or higher

Release Details

Full title	Topaz Gigapixel AI Pro
Version string	8.4.4
Publisher	Topaz Labs
Category	Photo Editing
Licence	Full version, all models included
Platform	Windows 10 and 11, 64-bit
Interface language	English
Archive size	2.14 GB
Mirrors	4 active, all reporting online
Listed	refreshed 7 weeks ago, checked again 40 minutes ago

Installation Notes

• Try several models on a crop before committing a batch. Model choice changes the result far more than any slider does.
• Enlarge before sharpening in your editor, not after. Sharpening first gives the model artefacts to amplify.
• Expect large output files. A six times enlargement of a modern camera file is a very large TIFF, so check your disk space before a batch.

Before You Start

Read the notes above before running setup. Most of the problems people write in about are covered there, and almost all of them come down to an older version of the same software still being installed, a graphics driver that predates the release, or a security suite quarantining part of the package mid install.

There is no archive password on this or any other entry in the library. If something you downloaded elsewhere claims to come from here and asks for one, it did not come from here. Nothing on this site is behind a survey, a shortener or a countdown either.

If the entry does not match what the page describes, say so on the contact page and include the version string from the release details table. Reports naming a specific version get checked first because the checker can reproduce them straight away.

Frequently Asked