JPEGmini Pro Tested: Pixel-Level Diff Tests & Comparisons

As a photographer, my websites are heavy on images. I strongly dislike images that look heavily compressed, with those tell-tale artifacts and degraded quality. But I also like fast page load times.

While developers argue over exactly how obsessive you need to be about site speed, the reality is simple: a fast website has no downside. Google rewards it, and your visitors expect it.

But raw image files carry a massive amount of redundant data. Some of it is hidden metadata — like camera settings, edited dates, and GPS coordinates. Some of it is optical data that human eyes simply cannot see. Stripping that redundant data out drastically reduces a file’s size. Multiply that size reduction across every graphic on a webpage, and the speed gains are massive.

I squeeze every byte of redundant data out of my files. Usually, I rely on traditional image optimization. This strips out hidden metadata using lossless algorithms, meaning the actual image quality isn’t touched.

You can push filesizes even lower using lossy compression, but there’s a catch. Lossy compression permanently discards optical information. If you get too aggressive, your images will look terrible. Exporting a JPEG from Photoshop at a quality setting of 100 looks flawless but creates a bloated file. Exporting at a quality setting of 20 creates a tiny file that looks like a pixelated mess. It’s not worth ruining your portfolio just to save a few kilobytes.

Optimization is always a balancing act. Because I’m highly protective of my image quality, I tend to be very cautious with lossy compression.

However, after running extensive pixel-level tests, I changed my workflow. I now use a specific lossy compression tool on my sites. It doesn’t just blindly crank up standard JPEG compression. Instead, it uses a proprietary algorithm to crunch files down dramatically without any visible hit to image quality.

The tool is JPEGmini. I have used it for years, originally reviewing older versions of the software. These tests are updated for JPEGmini Pro 3, which is still the current desktop generation — the builds on JPEGmini’s download page are 3.5.x. Version 3 brought a refreshed interface and HEIC support, and later updates added H.264/H.265 video compression, but the core image optimization engine is fundamentally the same.

JPEGmini currently offers a server-based option for developers to run custom optimization pipelines. However, I am focusing purely on the desktop software here.

For anyone landing here looking for that side of it: JPEGmini Photo Server is a Linux command-line application that companies install on their own servers, and it’s sold as an annual license rather than the one-time desktop purchase. It’s built for sites processing images at volume, not for individual photographers running a few hundred exports a week.

The JPEGmini Pro desktop app comes in two tiers. The standalone app sells for $59. For $89, you can buy a suite that includes the app alongside export plugins for Lightroom, Photoshop, and Capture One. There’s also a permanently free tier that processes up to 50 images a day using the same compression algorithm as the paid app, although it leaves out video, batch resize, and exporting to a separate folder. On top of that, a 7-day free trial unlocks the full app so you can test it on your own files before buying a license.

JPEGmini Pro 3

The desktop app is incredibly simple. It consists of a basic drop panel and a progress bar. You just drag individual images or entire nested folders onto the panel and let it run. It handles massive batches easily — I’ve pushed folders containing over 60,000 images through it at once with no problem.

Historically, JPEGmini only worked on JPEGs. With version 3, it now accepts HEIC files as well, automatically converting and optimizing them into standard JPEGs.

There is almost nothing to configure. You have zero control over the actual compression strength. You simply tell the software whether to overwrite your original files or export the optimized versions to a new folder.

Unlike standard JPEG compression, you can’t accidentally ruin a file by processing it twice. Once an image runs through JPEGmini, the software flags it. If you drop that same image back into the app, it simply skips it to prevent cumulative quality loss.

Your actual file size savings will vary. Images with simple details and flat tones compress more efficiently than highly detailed, noisy images. However, it is incredibly rare for me to run a JPEG through this software and not see a significant file size reduction.

JPEGmini Pro & Lightroom

If you buy the Pro Suite, you get a dedicated Lightroom export plugin. This is a massive workflow upgrade. You can bake JPEGmini directly into your existing export presets. It also integrates perfectly with Lightroom Publish Services — you just add it as a final step before the files are uploaded.

Results

I originally ran these tests using the older version of JPEGmini Pro. After testing version 3, the optical results and compression ratios remain identical.

I pushed a wide variety of subjects through the algorithm — from smooth gradients to razor-sharp, high-contrast architecture. The results are incredibly impressive. Toggle the sliders below. You simply cannot see a visual difference between the heavy original and the crunched file.

While the visual quality remains identical, the file size drops are staggering.

Image	KB	JPEGmini KB	Reduced by %
	765.2	325.8	57.4
	218.4	100.6	54.0
	459.7	277.7	39.6
	163.5	93.7	42.7
	319.4	212.7	33.4
	232.5	121.5	47.7
	336.6	208.7	38.0
	538.7	246.8	54.2
	197.2	75.7	61.6
	552.2	241.1	56.3

I wanted to see how far I had to push standard Photoshop compression to match these file sizes. I went back to the master RAW files and generated new 1024px JPEGs using Photoshop’s “Save for Web” dialog.

To match JPEGmini’s tiny file sizes, I had to drop Photoshop’s quality slider to anywhere between 60 and 77. At 77, an image still looks decent. But down at 60, aggressive banding and ugly artifacts destroy the image. JPEGmini is objectively vastly superior to standard JPEG compression.

Diff tests

Because the naked eye can’t spot the changes, I ran deep-level diff tests. This software analyzes the original file and the JPEGmini file down to the exact pixel, mapping out every single optical change.

In the images below, solid white represents untouched pixels. Solid black represents pixels that JPEGmini altered. This shows you exactly where the algorithm is working.

You can clearly see that the algorithm attacks hard edges and busy details, leaving smooth tonal areas completely alone. Because it hides the compression inside busy textures, the human eye can’t detect the loss. This completely avoids the ugly banding artifacts that ruin blue skies in standard JPEG compression.

JPEGmini vs traditional optimization

JPEGmini calls its process “optimization,” but it is technically a re-encoding tool. It actively compresses the optical data, but it completely ignores the hidden metadata. That means you can combine JPEGmini with traditional lossless optimization tools for maximum savings.

In the tests below, I ran files through JPEGmini, then ran them through ImageOptim, and finally tested a combination of both. ImageOptim was set to its most aggressive lossless setting, ensuring no further image degradation.

I exported these test JPEGs from Lightroom at 1024px with a quality setting of 85. (Note: larger, modern Retina images yield an even higher percentage reduction, but 1024px provides a solid, conservative baseline).

Image	KB	ImageOptim only KB	Reduced by %	JPEGmini only KB	Reduced by %	JPEGmini & ImageOptim KB	Reduced by %
	765.2	662.2	13.5	325.8	57.4	260.3	66.0
	218.4	183.5	16.0	100.6	54.0	72.1	67.0
	459.7	392.5	14.6	277.7	39.6	224.5	51.2
	163.5	139.2	14.9	93.7	42.7	73.1	55.3
	319.4	278.4	12.8	212.7	33.4	181.0	43.3
	232.5	200.6	13.7	121.5	47.7	94.5	59.4
	336.6	291.4	13.43	208.7	38.0	174.8	48.1
	538.7	467.3	13.3	246.8	54.2	197.0	63.4
	197.2	165.6	16.0	75.7	61.6	52.4	73.5
	552.2	477.2	13.6	241.1	56.3	184.8	66.5

Using both tools together provides the absolute lowest file size. Ideally, JPEGmini would build metadata stripping right into its app. But until they do, Mac users willing to use the command line can script ImageOptim and JPEGmini to run sequentially. Here is how to set that up.

One bonus for web developers: JPEGmini natively outputs Progressive JPEGs, which makes page loading feel much faster for the end user.

JPEGmini vs WebP

Since website speed is the primary goal, we have to compare JPEGmini against Google’s modern WebP format. I have tested WebP extensively before.

WebP promises much smaller files than JPEGs while retaining the same visual quality. While older browsers used to struggle with it, WebP is now universally supported. But how does it actually compare to a JPEGmini file?

Because WebP compression uses a sliding scale (just like standard JPEGs), I had to standardize the test. I used the Optimus cloud service to generate the WebP files, as it represents a standard WordPress workflow. I ensured both tests started from the exact same master file to avoid cumulative compression errors.

File Size Results: WebP is the clear winner here. The Optimus WebP files are significantly smaller than the JPEGmini files. They even beat out the files run through the combined JPEGmini + ImageOptim workflow.

Image	KB	JPEGmini only KB	Reduced by %	WebP KB	Reduced by %
	765.2	325.8	57.4	240.5	68.6
	218.4	100.6	54.0	30.6	86.0
	459.7	277.7	39.6	112.5	75.5
	163.5	93.7	42.7	23.4	85.7
	319.4	212.7	33.4	76.3	76.1
	232.5	121.5	47.7	37.4	83.9
	336.6	208.7	38.0	88.1	73.8
	538.7	246.8	54.2	165.1	69.4
	197.2	75.7	61.6	25.0	87.3
	552.2	241.1	56.3	138.9	74.9

Image Quality Results: JPEGmini wins on optical quality, though the margin is incredibly tight. On sharp architectural lines, the formats are visually identical. However, WebP struggles with subtle gradients and skies. In the Athens sunset shot below, JPEGmini holds the cloud detail perfectly, while the WebP version looks noticeably muddy.

Take a close look. The left is JPEGmini; the right is WebP.

Now look at the blurred, out-of-focus edges on the center tulip. The WebP version struggles slightly with the smooth tonal falloff.

JPEGmini & WordPress

Running your master images through JPEGmini before uploading them to WordPress sounds great, but there is a major trap. WordPress automatically generates smaller, derivative images (thumbnail, medium, large) upon upload. These auto-generated images are the ones WordPress actually serves to your visitors. And unfortunately, those derivatives completely bypass JPEGmini.

JPEGmini does not offer a native WordPress plugin. Right now, there are three practical workarounds:

• The Manual Route: Download all media via FTP, process it through the JPEGmini desktop app, and overwrite the server files. This is incredibly risky. You can easily break file permissions and ruin your media library.
• The Developer Route: Deploy a custom JPEGmini server. This requires significant technical skill and developer-level licensing.
• The Cloud Route: Use the EWWW Image Optimization Cloud plugin. JPEGmini is built directly into their cloud processing engine. It is the easiest method, but processing thousands of thumbnails can get expensive.

Wishlist

The core engine is nearly perfect, but the UI has room for improvement.

• Include metadata stripping: It would be incredibly convenient to handle lossy compression and lossless metadata removal in a single click without chaining it to ImageOptim.
• Dock functionality: On macOS, I want to drop files directly onto the closed dock icon. Currently, you have to open the app window to use the drop zone.

Conclusion

JPEGmini trims massive amounts of fat from my JPEGs without compromising optical quality. Since images make up roughly 75% of my page weight, crunching those files noticeably improves my load times.

I rely on it for my own sites. Every JPEG you see on Have Camera Will Travel has passed through JPEGmini (excluding the unedited diff tests above). Now that browser support has caught up, I also pair it with WebP and AVIF generation (JPEGmini can’t do those — I use other tools for that part) to strictly adhere to Google’s Core Web Vitals.

I highly advise against using JPEGmini on your master archive files. It is a lossy tool; once data is discarded, you can never get it back. But for web delivery, it is unmatched.

Download it from the JPEGmini website and test it on your own files — the free tier’s 50 images a day is more than enough to see what it does to your photos before you pay for anything. Shaving 20% to 50% off your image payloads lowers your bandwidth costs, speeds up your site, and keeps your portfolio looking sharp.