Gigapixel 8 soundly outperforms all other standalone upscalers on the market today. Furthermore, with the addition of new, albeit computationally-taxing, generative AI features, Gigapixel can produce results well beyond the reach of its rivals. Jump to Conclusion | Try Gigapixel for Free
REVIEW CONTINUES BELOW
GET GIGAPIXEL NOW
LIFETIME LICENSE
$99.99
30-DAY MONEY-BACK
Gigapixel Review – Contents
- What is Gigapixel 8
- What’s new in Gigapixel 8
- How much does Gigapixel cost?
- Download the Free Gigapixel Trial.
- Gigapixel Features
- Gigapixel Tests
- What is Gigapixel like to use
- Alternatives to Gigapixel
- Conclusion
- About the Author
What is Gigapixel 8
Gigapixel is an AI-powered image upscaling application for PC and Mac designed to enlarge your digital photos and pictures by up to 600%. As a standalone software application, you can run Gigapixel independently of any other software or as a plugin for Lightroom Classic and Photoshop.
But unlike traditional upscaling, which mindlessly turns one pixel into many, Gigapixel’s machine-learned AI algorithms read your image, anticipating and inserting new detail throughout the enlargement process. Thus, pictures enlarged with Gigapixel are not only bigger but often sharper and more detailed than the original image.
What’s new in Gigapixel 8
Gigapixel 8 features two new AI tools. The first is Recover, designed to restore and upscale low-resolution photos. The second, Redefine, uses generative AI to reconstruct a new picture in the original photo’s image with increased detail and superior image quality overall.
However, both of these features are computationally demanding. Although running these technologies locally on your computer is possible, you must exercise extreme patience. If patience is not an option, you can refer processing to Topaz Lab’s Cloud Renderer service. While this results in a much faster render, you must pay for the privilege.
Other additions include improved Face Recovery, a refined Comparison view, and a tweaked interface. Existing Gigapixel users should note that Gigapixel 8 is no better than Gigapixel 7 regarding standard upscaling. However, this is no bad thing as Gigapixel 7 was already best-in-class.
Overall, Gigapixel 8 is a considerable upgrade over Gigapixel 7, but only if you plan to exploit its new AI features. Upgrade to Gigapixel 8 Now
How much does Gigapixel cost?
A single Gigapixel 8 license will cost you a one-time payment of US$99.99 or $79.99 if you upgrade from a previous version. For comparison, Gigapixel’s main but much less capable rival, ON1 Resize AI, also costs US$99 but is often heavily discounted. Buy Gigapixel Now.
Meanwhile, 20 Cloud Rendering credits will cost you $5 ($0.25 per credit). However, credits are far cheaper when bought as a monthly plan. For instance, you can receive 80 monthly credits for US$9.99 or as much as 1400 credits for US$99.99. How much a render will cost you depends on your exported file size. Buy Cloud Render Credits Now
Download the Free Gigapixel Trial.
You can try a full-featured version of Gigapixel free for 30 days. No credit card is required. However, the upscaled copy of your photo will be watermarked. Try Gigapixel Now – For Free.
Gigapixel Features
Upscaling
Gigapixel is an upscaler that uses artificial intelligence to add additional detail to your image as it’s enlarged. To this end, Gigapixel features five AI models – Standard, High Fidelity, Low Res, Text & Shapes, and Art & CG. Each model has been optimized to suit different types of images of varying levels of quality. By default, Gigapixel will automatically choose which AI model it thinks will work best upon loading your image.
Alternatively, you can choose an AI model for yourself. The easiest way to do this is via Gigapixel’s comparison mode, which lets you preview and compare the effects of up to four different models simultaneously. Once you have chosen your AI model, you can apply additional adjustments to noise reduction, sharpening, and compression fixing.
Gigapixel will upscale images up to 600% of their original size. You can specify your output size as a multiplication factor, such as x4, or by using explicit dimensions in centimeters, inches, and pixels. When you’re happy with the result, you can export the upscaled copy of your image as a TIFF, PNG, or JPEG. Your original image is left intact.
Recover
Recover is an advanced AI model that uses Generative AI to upscale extremely low-resolution images (less than 1000px). Because Recover is computationally expensive, you can only preview its effect within a small moveable window. Furthermore, you must wait while the preview is refreshed each time you adjust Recover’s settings.
Once you’ve reached a result you’re happy with, you can render the image locally using your hardware or spend some credits and use Gigapixel’s Cloud Rendering for a faster result.
As with other generative AI tools, Recover is unpredictable because it produces a different result each time—even when applied to the same image. Furthermore, it takes an extremely long time to render a Recovered image locally, and there’s no guarantee you’ll like its first or second attempt. This is particularly frustrating given you’ve either waited an age for a local render or paid for a Cloud render.
But critically, there were many cases when I preferred the rendered-in-seconds output of Gigapixel’s standard AI models. Yet, I have no doubt that Recover will have its moments and produce the best possible result. Unfortunately, its unpredictable nature and speed issues make it challenging to know when.
Redefine
Redefine uses Generative AI to create a new, superior version of your original photo. Thus, Redefine works best on blurry or low-resolution images. Furthermore, you can use Redefine to reimagine your image. For example, you can add a text description to your image, and Redefine will add or substitute pictures accordingly.
Like other Generative AI tools, Redefine’s results are unpredictable and often unrepeatable. Due to its computationally taxing nature, rendering your Redefined image will cost you considerable time or a few dollars if you use Gigapixel’s fast Cloud Renderer. All this means you could spend a lot of time and money waiting for a result you never wanted.
Yet none of this is to say that Redefine isn’t capable as it might just produce the best possible result you could have hoped for. But it probably won’t, at least the first time.
Cloud Renderer
Gigapixel’s Cloud Renderer’s cloud-based processing is a faster alternative to rendering Reover and Redefine-process images with your computer’s hardware. How much faster depends on how powerful your PC or Mac is, but the difference might save you minutes or even hours.
Cloud Renderer is a premium option with a range of monthly subscriptions. Alternatively, you can pay as you go and buy credits as you need them for $0.25 each. The cost of upscaling with Cloud Renderer depends on the size of your output image. For instance, it will cost you six credits to upscale a picture to 12 megapixels, whereas a 60 megapixel upscale will cost you 20 credits.
Using Cloud Renderer is exceptionally easy. After applying to Recover or Redefine, you can export your processed image for local render or export to the Cloud Renderer. Your image or images are then added to a queue and appear in your Download folder (PC Version tested) once processed. Your credit balance is displayed at the top right of Gigapixel’s interface, and adding credits is easy enough.
Cloud Renderer is a convenient and well-implemented feature in itself. However, its value is determined by how fast your local hardware is and how much you like Redefine and Recover. Explore Cloud Renderer Prices.
Face Recovery
Gigapixel has long been my favorite upscaler for enlarging portraits. While most upscalers tend to over-process, Gigapixel’s lighter touch preserves natural features and skin tones. Nonetheless, Gigapixel also has a Face Recovery feature, which enables you to dial back the processing further, and new to Gigapixel 8 is Gen 2 Face Recovery.
Ironically, I’ve found Face Recovery to be quite destructive. For the most part, Gigapixel gets faces right, and switching on Face Recovery tends to wash away details and produce waxy skin tones. That said, I have found rare occasions where applying just a small amount of Face Recovery has improved an image.
Therefore, while I don’t use Face Recovery much, I like having it for those rare occasions when I’d like to dial back the processing a little.
Crop and Gamma
Including Crop and Gamme tools reduces Gigapixel’s dependence on separate photo editing applications. As a result, you’re less likely to spend time bouncing your image between different software applications.
Gigapixel Tests
Test Method
I begin by creating a copy of a full-size reference image. Next, I shrink the copy four-fold and then, using an upscaler, enlarge the image by 400%, thus restoring the copy to its original size. Finally, I compare the upscaled copy with the original to see how accurate the enlargement is.
Test 1
Gigapixel succeeds in creating an upscaled image that is faithful to the original and improves it.
Test 2
Gigapixel 8 is the best upscaler for enlarging portraits. Alternatives to Gigapixel, such as Adobe Super Resolution and ON1 Resize AI, tend to overprocess, resulting in unnatural-looking features and skin tones.
Test 3
While Gigapixel fails to produce a replica of this complex image, it gets much closer than any other upscaler and is entirely passable when viewed at practical distances.
What is Gigapixel like to use
Gigapixel 8’s interface is exceptionally straightforward and stocked with well-implemented features. Moreover, its automatic modes are highly reliable and likely to deliver near-optimum results without any effort on your part. If you prefer to make your own adjustments, Gigapixel’s excellent comparison view makes it easy to choose the most suitable AI model.
Moreover, Gigapixel has never been faster. It responds to your adjustments in real-time, and the wait, while it updates the preview to match your new adjustments, has never been so brief—unless you are using Recover or Redefine.
Both Redifine and Recover exploit diffusion-powered generative AI, a technology that still demands more than most computers can handle. Therefore, Gigapixel will take 30-60 seconds to preview only a tiny part of the image. If you want to render the image locally, you could be waiting minutes or hours, depending on the size of the image and the power of your computer’s graphics card.
For example, rendering a Redefined 2-megapixel photo using a 16GB AMD RX 7800 XT GPU took me about 5 minutes. I also used Redefine to deblur a 20-megapixel file and canceled the render after 90 minutes as the progress bar was still on 0%. Making matters worse, you cannot use Gigapixel 8 while a local render takes place.
Therefore, you have two solutions: buying a monster GPU, such as the Nvidia RTX 4090, or pumping credits into Cloud Renderer. I dare say Cloud Renderer offers better value, and you can still use Gigapixel while the render takes place in the background.
Alternatives to Gigapixel
Gigapixel vs Topaz Photo AI
Gigapixel and Topaz Photo AI are equally excellent when upscaling good-quality images. However, Topaz Photo AI also includes noise reduction and blur-correcting sharpening. Best of all, you can combine these features to upscale blurred and noise-ridden images.
With Recover and Redifine, Gigapixel 8 has closed the gap on Topaz Photo AI regarding upscaling poor-quality photos. However, Topaz Photo AI’s technology is more reliable, predictable, vastly faster, and runs perfectly without the Cloud.
Unfortunately, at US$199, Topaz Photo AI costs twice as much as Gigapixel. So, if you’re only upscaling sharp and noise-free images, you would be sensible to save your money and buy Gigapixel. But if you want the best upscaler for all conditions, get Topaz Photo AI.
Read my Topaz Photo AI Review for more information, comparisons, and your free trial.
Gigapixel AI vs Lightroom Super Resolution
Gigapixel is superior to Lightroom Super Resolution for two reasons. First, unless you’re willing to endure a simple but clumsy workaround, Lightroom Super Resolution is limited to 200% upscales compared to Gigapixel’s maximum of 600%.
Second, my tests have found that Gigapixel is better than Lightroom Super Resolution when upscaling complex scenery and portraits. Nonetheless, if you are a Lightroom subscriber with infrequent upscaling needs, you might find Lightroom Super Resolution good enough. But, regarding image quality, Gigapixel is leagues ahead.
Read Lightroom Review for more information, samples, and comparisons.
Gigapixel vs ON1 Resize AI
Gigapixel is a better upscaler than ON1 Resize AI, particularly when enlarging portraits. At one time, I would have considered ON1 Resize AI because it was so much faster than Gigapixel. But now, Gigapixel is, at least, fast enough.
ON1 Resize AI has a few print-friendly features, such as presets optimized for specific types of printers and paper and a canvas wrap tool. However, regarding upscaling, ON1 Resize is soundly beaten by Gigapixel.
Read my ON1 Resize AI Review for more information, comparisons, and a free trial.
Get Discounts on Photo Editing Software
Subscribe to my weekly newsletter and be notified of deals and discounts on photography software from ON1, Adobe, Luminar, and more. Spam Promise: Just one email a week, and there’s an unsubscribe link on every email.
Conclusion
Gigapixel is by some margin the best standalone upscaler in photography due to its superior image quality and outstanding usability. Regarding upscaling good-quality images, Gigapixel is matched only by its stablemate, Topaz Photo AI.
Moreover, thanks to generative AI, Gigapixel now has the potential to correct and upscale noise-ridden and blurred images and even transform your photo’s content. Unfortunately, while these tools have outstanding potential, they remain somewhat unpractical.
The fundamental problem with Generative AI tools such as Recover and Redefine is their unpredictability. You can never predict the quality or nature of the result or even repeat it – even when working on the same photo. Therefore, it may take you many tries to get the result you are looking for, and that’s a big problem.
Recover and Redefine require a lot of processing power. As a result, you can only preview the effects of your adjustments through a small window, which can take a minute to render. Moreover, rendering an entire image using your computer’s hardware can take minutes or hours. Worse still, Gigapixel 8 will remain inoperable until the render is complete.
For this reason, Gigapixel 8 now offers Cloud Renderer. You can leverage Topaz Lab’s powerful Cloud technology in exchange for your hard-earned credits. Not only is this faster, but you can continue to use Gigapixel while the render takes place. But, whether you choose to wait or opt to pay, Recover and Redefine’s trial-and-error workflow feels entirely unpractical.
In contrast, Topaz Photo AI uses comparatively lightning-fast technologies to upscale poor-quality images with predictably good results. While it is possible that Gigapixel’s generative AI could do an even better job, it’s impossible to tell how many hours or dollars it would cost to find out.
Yet, despite their impracticalities, these new technologies are bold, forward-thinking additions that subtract nothing from an already outstanding product. Therefore, Gigapixel remains, by far, the best upscaler for enlarging good-quality images. But if you want to upscale blurred and noise-ridden photos, Topaz Photo AI is the way to go. Try Gigapixel – Free | Try Topaz Photo AI – Free
Subscribe to my weekly newsletter and receive deals and discounts on Gigapixel AI and other photography software and gear. Subscribe Now.
GET GIGAPIXEL NOW
LIFETIME LICENSE
$99.99
30-DAY MONEY-BACK
OR TRY GIGAPIXEL FOR FREE
About the Author
Further Reading
Topaz Photo AI Review – Now with Super Focus
Gigapixel Review – Gigapixel 8 Pushes the Envelope
Topaz Denoise AI Review – A Flawed Gem.
Sharpen AI vs Denoise AI – Which Should you Buy First?
OM Workspace Review – AI Noise Reduction Tested and Compared
Topaz Video AI Review – Amazing Features, Flawed Performance.
