Aiarty Image Enhancer 3.14 — Mac version

The Quadro-Detail generator is where the one-click crowd get filtered out. It doesn't just sharpen edges—it rebuilds skin pores, hair strands and weave patterns from whatever mess you've fed it. Getting that right on a noisy scan from 2004 takes a different kind of tool.

What experienced hands will poke at first

• Output to 32K, which means you can upscale a tight crop into something printable at gallery size without the plastic-smooth look of old-school interpolation.
• AVIF and WebP support alongside the usual TIFF and PNG—useful if you're feeding a modern web pipeline and want to keep file sizes sensible.
• Lightroom Classic workflow improvements baked into version 3.14. Send a batch over, let the deep learning models do the denoising and face reconstruction, then carry on editing without leaving your catalogue.
• Manual control over the generation model. The app can guess which textures to rebuild, but you can tell it to prioritise fabric over skin, or vice versa, when the auto result looks overcooked.

Batch TIFFs without a browser crash

The desktop-only approach means it'll chew through a batch of high-res TIFFs without a browser tab crashing halfway through. It runs back to macOS 11, so an older Intel machine with a decent GPU still gets a look-in. For archive restoration—old family photos, scanned negatives, product shots shot in bad light—the combination of deblurring and texture-aware upscaling pulls back detail that simpler sharpeners smooth into oblivion.

Hand-off frames that need deep reconstruction

Aiarty isn't a library organiser, and it doesn't pretend to be. You do your culling and colour work elsewhere, then hand off the frames that need deep reconstruction. The EXIF metadata access added in 3.14 means you can check what ISO a noisy shot was pushed to without hopping back to Lightroom—small thing, but it saves a trip.

Drag a folder of RAW files straight from Finder and the app reads them without a separate converter. That matters when you're pulling old .CR2 files off a dusty hard drive and you don't want to rebuild a whole Adobe pipeline just to see if the frames are salvageable. For web work, export to AVIF and drop the result into your site's assets folder. No intermediate conversion step, no quality loss from a second compression pass.

On the macOS side, it respects the system colour profile and hands off to Preview for quick checks. If you've set up a hot folder with an Automator script, Aiarty will pick up new images as they land—useful for a shared scanning station where volunteers dump old prints and want them cleaned up without learning the app themselves.

Leaner upscaling on Intel hardware

Topaz Gigapixel AI has been the go-to upscaler for years, and it still leads on batch consistency. But it's heavier on the GPU and more fiddly with older macOS versions. Aiarty runs leaner on Intel hardware and doesn't demand the latest Metal feature set, so a 2015 iMac with a bumped-up graphics card still gets usable throughput. Where Topaz sometimes over-smooths low-res faces, Aiarty's Quadro-Detail generator takes more risks on texture reconstruction—sometimes it wins, sometimes it hallucinates a bit of eyebrow that wasn't there.

HitPaw Video Enhancer comes up in searches because people assume video and photo enhancement are the same category. They're not. HitPaw does frame interpolation and video upscaling competently, but it's a different beast—no RAW support, no EXIF tools, and the face model is tuned for moving images, not the kind of fine static detail a photographer cares about. If you're restoring a single frame from a video grab, Aiarty gives you more control over what gets rebuilt.

Upscayl is the open-source contender. It does basic upscaling well and costs nothing, but it lacks the texture-aware generation that Aiarty brings to fabric, skin and hair. For a quick 2x upscale of a clean image, Upscayl is fine. For a damaged scan where you need denoising and detail reconstruction in one pass, Aiarty pulls ahead.

Offline use with no activation server

Does it phone home? The TNT build works entirely offline. No account, no activation server, no background chatter. If your repair-café Mac lives on a local network with no internet, the app won't complain.

Will it run on an old Intel Mac? Yes, as long as you're on macOS 11 or later. On machines without a discrete GPU, upscaling to 32K will be slow—expect to leave it overnight for a batch of large TIFFs. For web-resolution work, it's perfectly usable on integrated graphics.

Can I batch-process different file types together? You can. Drop a mix of JPEGs, RAWs and PNGs into the queue and Aiarty handles them in one go. The output format is set globally, though—you can't assign AVIF to some and TIFF to others in the same batch.

What's the catch with Gatekeeper? The disk image is Open Gatekeeper friendly, so a right-click and Open usually does it. If not, the Terminal bypass is one line and you're done. Occasionally an Intel Mac wants Rosetta enabled for the first launch; ticking that box in Get Info sorts it.

• macOS 11 or later
• Intel or Apple Silicon Mac