Forem: Luca Sammarco

Optimize Images for Shopify: Free Guide [2026]

Luca Sammarco — Wed, 15 Apr 2026 18:26:27 +0000

Slow Shopify stores lose customers. Research consistently shows that a one-second delay in page load time can reduce conversions by up to 7%. For most Shopify merchants, oversized product images are the single biggest culprit — and the easiest fix.

This guide shows you exactly how to optimize images for Shopify before you upload them, using free browser-based tools. No apps, no monthly fees, no server uploads.

Why image optimization matters for Shopify

Shopify is a well-built platform, but it cannot fix problems you create at the source. When you upload a 4MB product photo, Shopify stores that file and generates smaller versions from it. All those generated sizes are derived from your bloated original.

Problems:

Slower FCP — Core Web Vitals affect your search ranking
Higher bounce rates on mobile — pages taking 3+ seconds get abandoned
Wasted storage — unoptimized images compound across hundreds of SKUs

Shopify image size requirements

Image Type	Dimensions	Target Size
Product images	2048 x 2048 px	Under 200KB
Collection images	1024 x 1024 px	Under 150KB
Slideshow / Banner	1920 x 1080 px	Under 300KB
Logo	250 x 250 px max	—
Favicon	32 x 32 px	—
Blog post images	1200 x 628 px	Under 200KB

The 3-step workflow (before upload)

Takes under 2 minutes per batch. Everything runs in your browser.

Step 1: Resize — Use SammaPix ResizePack to resize to exact dimensions. Product images 2048x2048, banners 1920x1080.

Step 2: Compress — Use SammaPix Compress at 80% quality. The sweet spot: human eye cannot detect quality loss, but file sizes drop dramatically. Batch up to 20 images.

Step 3: Convert to WebP — Use SammaPix WebP. Shopify Online Store 2.0 supports WebP natively. WebP is 25-35% smaller than JPEG at the same quality.

Real benchmark data

Image	Before	After	Savings
Product photo (white bg)	4.2 MB	180 KB	96%
Lifestyle shot (model)	3.8 MB	210 KB	94%
Model on white bg	2.1 MB	95 KB	95%
Collection banner	2.6 MB	165 KB	94%
Blog hero	1.9 MB	140 KB	93%

A typical Shopify product page with 4-6 images goes from 12-18MB to under 1MB.

Shopify-specific tips

Shopify generates multiple sizes from your original. All variants derive from the uploaded file. Optimize before upload = every generated size is proportionally smaller.

Filename matters. IMG_3847.jpg gives Google nothing. mens-leather-wallet-brown-bifold.jpg is SEO gold.

Use square product images. 2048x2048 (1:1) renders consistently across all themes. Non-square often gets cropped unpredictably.

SammaPix vs. Shopify optimization apps

Paid apps (TinyIMG, Crush.pics): $4-40/month. Compress AFTER upload via Shopify API. Variants may already exist in CDN cache unoptimized. $240/year indefinitely.

SammaPix (free): Optimize locally before upload. Free forever. Works for any platform. Shopify generates all variants from already-optimized source.

For ongoing uploads, manual pre-upload is faster and cheaper. For large legacy catalogs, run a plugin once, then switch to pre-upload going forward.

Check your current speed

Run PageSpeed Insights on your homepage + a product page
Look for "Serve images in next-gen formats" and "Properly size images"
Note the savings in KB next to each image
Prioritize biggest savings first

Most unoptimized Shopify stores score 40-65 on mobile. After the workflow, scores reach 75-90.

Checklist

Resize to correct dimensions (product: 2048x2048, banner: 1920x1080)
Compress to 80% quality, under 200KB for product images
Convert to WebP if your theme supports Online Store 2.0
Use descriptive filenames
Write descriptive alt text after uploading

All three steps are free at SammaPix — browser-based, batch up to 20 images.

Originally published on sammapix.com/blog

WordPress Image Optimization: Why You Should Compress Before Upload (2026 Guide)

Luca Sammarco — Tue, 14 Apr 2026 07:55:12 +0000

The credit-based plugin trap.

Most WordPress image optimization plugins use a credit system. You install the plugin, it compresses your images on their servers, and you pay per image or per megabyte. ShortPixel gives you 100 images per month free, then you pay per credit. A single blog post with 5 images burns through your monthly allowance in a few weeks. Imagify gives you 20 megabytes per month free, which is roughly 10 photos from a modern phone. One product shoot and you are done for the month. Smush Pro costs 49 dollars per month as part of the WPMU DEV bundle. The free version caps compression at 5 megabytes per image and strips lossy compression entirely. WP-Optimize has a limited free tier with basic compression and advanced features locked behind a paid plan.

The frustration is real. One Reddit thread with 46 upvotes and 71 comments is titled "I am so tired of Credit Based Image Optimizers." Another comment with 30 upvotes puts it bluntly: "I don't understand why people put image optimizers on their websites instead of optimizing before upload."

The core problem is simple. You are paying a monthly fee for something you can do once, for free, before you ever upload the image. Every image that hits your WordPress media library should already be optimized. The plugin becomes unnecessary.

The "compress before upload" workflow.

Here is the step-by-step process. It takes about 30 seconds per batch of images. First, export your images as usual — JPG from your camera, PNG from Canva or Figma, screenshots from your phone. Second, open SammaPix Compress — no account needed, runs entirely in your browser. Third, drop all images at once — batch up to 20 images in a single drop. Fourth, set quality to 80 percent — this is the sweet spot for web. Visually identical, dramatically smaller files. Fifth, download the compressed images — they are ready for WordPress. Sixth, upload to WordPress — your images are already optimized. No plugin processing, no credits spent, no waiting.

Optional bonus step: convert to WebP first using the SammaPix WebP converter for an additional 25 to 35 percent file size reduction on top of the compression.

Everything happens client-side. Your images never leave your browser. There is no upload queue, no server processing, and no file size limits beyond what your browser can handle.

Real results — 3 test images before and after.

We tested three typical WordPress images through SammaPix Compress at 80 percent quality, then converted the results to WebP. A product photo from an iPhone 15 went from 4.2 megabytes to 420 kilobytes after compression, and down to 310 kilobytes after WebP conversion — a 93 percent total savings. A blog header exported from Canva went from 1.8 megabytes to 195 kilobytes after compression, and down to 145 kilobytes after WebP — 92 percent savings. A team photo from a DSLR went from 8.7 megabytes to 680 kilobytes after compression, and down to 490 kilobytes after WebP — 94 percent savings.

That product photo went from 4.2 megabytes to 310 kilobytes. On a blog page with 5 similar images, that is the difference between loading 21 megabytes of images and loading 1.5 megabytes. Your visitors notice that difference, and so does Google.

Results vary depending on image content and complexity. Photos with lots of detail compress less than simple graphics. These are actual test results, not guarantees.

What about WebP? Should you convert too?

Short answer: yes, if your WordPress is version 5.8 or later, which is virtually every active WordPress site in 2026. WebP is a modern image format developed by Google that delivers 25 to 35 percent smaller files than optimized JPEG with no visible quality difference.

WebP has 97 percent browser support as of 2026 — every major browser including Safari, Chrome, Firefox, and Edge. WordPress has supported WebP uploads natively since version 5.8, released July 2021. WebP saves 25 to 35 percent more than an already-optimized JPEG — that is free performance on top of compression. The only browsers that do not support WebP are Internet Explorer, which has been discontinued, and very old Safari versions.

The ideal workflow is: compress with SammaPix Compress, then convert to WebP with the WebP converter, then upload to WordPress. Two steps, zero cost, maximum compression.

You can also go directly from JPG or PNG to WebP using the JPG to WebP or PNG to WebP converters, which handle compression and format conversion in a single step.

But what about lazy loading and CDN?

A common objection: "I do not need to compress images because my CDN handles it" or "lazy loading takes care of performance." This is a misunderstanding of what each technology actually does.

Lazy loading delays when an image loads, but does not reduce its file size. Your server still stores a 4 megabyte file, and the visitor still downloads 4 megabytes — just later in the page scroll.

A CDN, or Content Delivery Network, caches your images on edge servers closer to the visitor, which reduces latency. But it serves exactly what you uploaded. A CDN does not compress your images — it delivers your 4 megabyte file faster from a closer location.

Image compression is the only thing that actually reduces the bytes transferred. A 4 megabyte image compressed to 400 kilobytes saves 3.6 megabytes of bandwidth on every single page load, regardless of lazy loading or CDN.

The best approach is all three together: compress before upload plus lazy loading plus CDN. Each solves a different problem. Compression reduces file size, lazy loading defers non-critical images, and CDN reduces latency. Skipping compression and relying on the other two is like putting racing tires on a truck — it helps, but you are still hauling unnecessary weight.

When you DO need a plugin (and when you don't).

Being honest: there are situations where a WordPress image optimization plugin makes sense. If you have 10,000 or more existing unoptimized images, a plugin can batch-process your entire media library retroactively. Running ShortPixel once to optimize your backlog, then canceling, is a valid strategy. If non-technical editors upload images daily, a plugin acts as a safety net to catch unoptimized uploads. If you need automatic WebP conversion on the fly, some plugins can serve WebP to supported browsers and fall back to JPEG for older ones. This is less relevant in 2026 with 97 percent WebP support, but still useful for edge cases.

For new images going forward, compress before upload is always better. It is free, does not add a plugin dependency to your site, does not require server resources, and gives you full control over quality settings.

The sweet spot for most WordPress sites: use SammaPix to compress every new image before upload, and if you have a large backlog of unoptimized images, run a plugin once to clean them up. Then uninstall the plugin. You do not need it running permanently.

Frequently asked questions.

Do I still need an image optimization plugin if I compress before upload? For new images, no. The plugin becomes redundant if you compress before uploading. For existing unoptimized images already on your server, a one-time plugin run can help batch-process them retroactively. After that, you can uninstall the plugin.

What quality setting should I use for WordPress? 80 percent is the sweet spot for most photos. It reduces file size by around 90 percent with no visible quality loss on screen. Logos and graphics with text should use higher quality, 90 percent or above, or stay as PNG to preserve sharp edges.

Should I upload WebP or JPEG to WordPress? WebP if your site uses WordPress 5.8 or later, which is every modern WordPress installation. WebP saves 25 to 35 percent more than optimized JPEG with no visible quality loss. WordPress has supported WebP natively since version 5.8.

How many images can I compress at once with SammaPix? 20 images per batch on the free tier, 500 on Pro. All processing happens in your browser — no upload to servers, so your images stay private and compression is instant.

Will compressing images affect my SEO? Positively. Smaller images mean faster page loads, which means better Core Web Vitals scores and higher Google rankings. Google has confirmed page speed is a ranking factor. A page that loads in 2 seconds instead of 5 seconds will rank better, all else being equal.

Originally published at sammapix.com

Try it free: SammaPix — 27 browser-based image tools. Compress, resize, convert, remove background, and more. Everything runs in your browser, nothing uploaded.

Batch Rename Photos with AI: SEO-Friendly Filenames (2026)

Luca Sammarco — Mon, 13 Apr 2026 14:02:36 +0000

Why image filenames matter for SEO (and why you are probably getting it wrong)

Google Image Search is one of the highest-traffic sources for photographers, e-commerce sites, and content creators. Unlike web pages, search engines cannot read image content directly. They use multiple signals to understand what is in a photo:

Filename: "woman-jogging-park-morning.jpg" is more informative than "IMG_4521.jpg"
Alt text: the HTML alt attribute describing the image
Page context: surrounding text and headings
Image metadata: EXIF data, dimensions, file size

Filenames are one of the few signals you completely control. According to Google's own documentation , descriptive filenames improve image discoverability. A photo of coffee equipment titled "espresso-machine-gaggia-classic.jpg" is more likely to appear when someone searches "gaggia espresso machine" than the same image named "photo.jpg".

The problem: most photographers and content creators use camera defaults or generic names. A typical DSLR produces 5000+ photos per project. Renaming them manually would take days. That is where AI comes in.

How AI batch renaming works

Modern AI models like Google Gemini can "see" images and describe them accurately. AI batch renaming takes advantage of this capability to generate descriptive names automatically.

The process

You select a batch of photos and send them to the AI renamer. For each image, the AI:

Analyzes the visual content (objects, people, scenery, activities)
Identifies dominant subjects and contextual details
Generates a concise, descriptive filename
Applies keyword-rich naming conventions for SEO
Proposes the new filename for your approval

The entire batch processes in seconds to minutes, depending on quantity and internet speed. You can accept all suggestions, edit individual names, or reject and try again.

What makes a good AI-generated filename?

Not all AI-generated names are equal. The best filenames are:

Concise: 3-6 words max. Long filenames are harder to read and don't add SEO value after 50 characters.
Keyword-rich: include searchable terms that describe the photo (e.g., "red-fox-snow-forest" instead of "animal-in-white-stuff").
Hyphenated: use hyphens between words. Dashes help search engines parse word boundaries. Never use underscores for SEO filenames.
Lowercase: all lowercase. Uppercase doesn't hurt, but lowercase is convention and cleaner.
Accurate: no misleading or keyword-stuffed names. "seagull-flying-beach" is better than "sexy-birdwatching-vacation-photos-beach-free-download".

Keyword-rich filenames improve search visibility for your images - Photo by Merakist on Unsplash

Real-world examples: before and after

Here is what AI batch renaming looks like in practice:

Original filename |
AI-generated name |
SEO improvement |

IMG_5381.jpg |
woman-hiking-mountain-trail.jpg |
Searchable for hiking, woman, mountain |

DSC_2847.JPG |
golden-retriever-playing-beach-sunset.jpg |
Strong for dog breed + location + time |

PHOTO_20260315.jpg |
fresh-salad-with-tomato-cucumber-feta.jpg |
Ranks for food photography queries |

pic1.jpg |
espresso-machine-gaggia-classic-counter.jpg |
Targets specific product + context |

vacation_photo_8.jpg |
piazza-san-marco-venice-italy-crowded.jpg |
Local SEO + landmark name |

How to batch rename photos with SammaPix AI Rename

The SammaPix AI Rename tool makes batch renaming simple. No plugins, no desktop software, no complex setup. Here is the workflow:

Step 1 - Select your photos

Open AI Rename and drag a folder of photos onto the upload area. You can select 5–200+ images at once depending on your plan. The tool accepts JPG, PNG, WebP, GIF, and HEIC formats. Mixed formats in a single batch work perfectly.

Step 2 - Configure naming style (optional)

Choose how descriptive you want the names:

Short (3 words): "woman-jogging-park.jpg"
Medium (5 words): "woman-jogging-in-park-morning-light.jpg"
Detailed (7+ words): "athletic-woman-jogging-on-forest-trail-at-sunrise.jpg"

Step 3 - AI generates suggestions

The AI analyzes each photo and proposes SEO-friendly names. The process takes 1–5 seconds per image depending on file size and internet speed. You see a preview of the new names side-by-side with your originals.

Step 4 - Review and edit

Accept names as-is, edit individual suggestions, or reject and regenerate. You have full control. If the AI suggests something inaccurate, change it in place. This takes seconds per image.

Step 5 - Download renamed files

Once you approve all names, download the renamed photos as individual files or as a ZIP archive. The original files stay in their original location. You get a new set with the SEO-friendly names applied.

Use cases where batch renaming with AI saves hours

Photographers and content creators

After a shoot, a photographer might have 500+ photos. Renaming them manually would take 8+ hours. AI Rename does it in 5–10 minutes, with SEO-optimized names ready to upload to stock photo sites (Unsplash, Pexels, Shutterstock) immediately.

E-commerce teams

Product photos with generic names like "IMG_0001" hurt your image search visibility. Rename to "red-leather-handbag-large-capacity" and you rank for product-specific searches. Batch process entire product catalogs in minutes.

Blog and content websites

Each blog post uses 5–20 images. Descriptive filenames improve image search traffic and on-page SEO. Instead of uploading "screenshot1.png", use "wordpress-woocommerce-product-page-setup.png".

Travel and lifestyle blogs

Geographic data in filenames helps local SEO. "eiffel-tower-sunset-paris" ranks better than "vacation_pic_2.jpg". Batch rename 100+ travel photos with location and activity keywords instantly.

AI batch renaming scales from single photos to thousands - Photo by Unsplash on Unsplash

Complementary tools: combine renaming with other optimizations

AI Rename is powerful on its own, but it works best as part of a complete image optimization workflow. After renaming, consider:

Compress images for web

SEO-friendly filenames are worthless if your images take 10 seconds to load. After renaming, use SammaPix Compress to reduce file sizes by 50–80% without visible quality loss. Smaller files = faster loading = better SEO rankings and user experience.

Convert to WebP for modern browsers

After renaming and compressing, convert images to WebP format for an additional 25–35% file size reduction. The SammaPix WebP converter maintains all your new filenames while converting the format.

Remove EXIF data for privacy

Before uploading renamed photos, consider removing EXIF metadata (GPS coordinates, camera model, timestamps). Good filenames improve SEO, but hidden location data is a privacy risk. Strip EXIF, keep filenames, publish with confidence.

FAQ

What is batch renaming photos with AI?

Batch renaming with AI automatically renames hundreds of photos at once using artificial intelligence. Instead of naming each file manually, AI analyzes the image content and generates descriptive, keyword-rich filenames that describe what is in each photo. This works for any image type- portraits, landscapes, products, food, animals, etc.

Why does image filename matter for SEO?

Image filenames are a direct ranking signal for Google Image Search. A descriptive filename like "woman-running-forest.jpg" ranks better than "IMG_0001.jpg". Google uses filenames to understand image content, and keyword-rich filenames increase the chances your images appear in relevant search results. This is one of the easiest, highest-leverage SEO improvements available.

How does SammaPix AI Rename work?

SammaPix AI Rename uses Google Gemini 1.5 Flash to analyze each image and generate a descriptive filename. The AI examines the visual content, identifies objects, people, scenes, and activities, then creates a concise, SEO-friendly name. Batch mode processes hundreds of photos at once. All processing happens in your browser- your images never leave your device or are stored on any server.

Can I batch rename photos without uploading them?

Your full-resolution images never leave your device or are stored on a server. AI Rename sends only a small thumbnail to Google Gemini for analysis. The analysis happens in seconds, and the thumbnail is discarded immediately. Your original image files are never uploaded, stored, or retained.

Is batch renaming with AI free?

SammaPix AI Rename is free for up to 5 renames per day on the free plan. For unlimited batch renaming and larger batches, upgrade to SammaPix Pro for $7/month. Pro gives you 200 AI renames per day, batch processing for up to 100 images at once, and ZIP download support.

What file formats does AI Rename support?

SammaPix AI Rename supports JPG, PNG, WebP, GIF, and HEIC formats. You can rename photos from iPhones (HEIC), DSLRs (after RAW conversion to JPG), smartphones, and web images all in the same batch. Mixed formats work perfectly.

Originally published at sammapix.com

Try it free: SammaPix — 27 browser-based image tools. Compress, resize, convert, remove background, and more. Everything runs in your browser, nothing uploaded.

Batch Compress Images Free — No Signup, No Upload to Server (2026)

Luca Sammarco — Mon, 13 Apr 2026 14:02:25 +0000

The problem with server-based image compression.

When you use a tool like TinyPNG, iLoveIMG, or Compressor dot io, your images are uploaded to their servers. The compression happens on their infrastructure, and the result is sent back to your browser for download. This round trip introduces three categories of problems.

Privacy risk. Your images temporarily exist on a third-party server. For personal photos, client work, medical images, legal documents, or any sensitive content, this is a real risk. Even if the service claims to delete files after processing, you are trusting their infrastructure, their employees, their data retention policies, and their ability to prevent breaches. According to IBM's 2025 Cost of a Data Breach Report, the average data breach costs 4.88 million dollars.

Speed bottleneck. Uploading a batch of 20 images at 3 megabytes each means transferring 60 megabytes upstream. On a typical home connection with 10 megabits per second upload speed, that is roughly 48 seconds just for the upload, before any compression even starts. Then the results need to be downloaded. With browser-based compression, the same batch processes in 5 to 15 seconds with zero network transfer.

Artificial limits. Server-based tools impose limits because processing costs them money. TinyPNG allows 500 free compressions per month. iLoveIMG limits batch size to 15 files. These restrictions exist purely because of server costs, not because of any technical limitation in image compression itself.

How browser-based image compression works.

Modern web browsers are powerful enough to compress images locally without sending them anywhere. This is not a compromise. It is how image compression should work. The technology stack involves three layers.

Layer 1: The Canvas API. The HTML5 Canvas API provides a toBlob method that can encode image data into JPEG, PNG, or WebP format at a specified quality level. When you load an image onto a canvas and export it at quality 0.8, the browser's native image encoder handles the compression. This is the same encoder the browser uses to render web pages, so it is highly optimized.

Layer 2: The browser-image-compression library. Libraries like browser-image-compression, used by SammaPix, build on the Canvas API to add intelligent features: target file size optimization, progressive quality reduction, EXIF preservation options, and Web Worker support for non-blocking compression.

Layer 3: OffscreenCanvas and Web Workers. The OffscreenCanvas API allows canvas operations to run in a Web Worker thread, separate from the main UI thread. This means image compression can happen in the background while you continue interacting with the page. Chrome, Firefox, and Edge all support OffscreenCanvas, with Safari adding support in Safari 16.4.

Quality comparison: browser versus server. A common misconception is that server-side compression is significantly better. In reality, modern browser engines are highly optimized. Browser-based tools achieve compression ratios within 5 to 10 percent of server-side tools like MozJPEG or libvips. The difference is imperceptible in virtually all web use cases.

Tool comparison: TinyPNG versus Squoosh versus iLoveIMG versus SammaPix.

TinyPNG uses server-side processing, handles up to 20 files per batch, limits to 500 images per month, and has a 5 megabyte file size limit. Your files are uploaded to their servers. Quality is automatic with no manual control.

Squoosh is browser-based and fully private, but processes only one file at a time with no batch support. It offers excellent manual quality control with multiple output formats including AVIF.

iLoveIMG uses server-side processing, handles up to 15 files per batch, with limited monthly batches. Your files are uploaded to their servers. Quality is automatic only.

SammaPix is browser-based and fully private. It handles 20 files per batch on the free tier and 500 on Pro. File size limit is 20 megabytes free, 50 megabytes Pro. It offers a full manual quality slider with JPEG, PNG, and WebP output. It includes 27 additional tools beyond compression.

The key takeaway: Squoosh and SammaPix are the only fully browser-based options. Squoosh processes one image at a time, which makes it impractical for batch workflows. SammaPix processes up to 20 images at once on the free tier.

Why no-signup matters more than you think.

Requiring an account for image compression is a dark pattern. Image compression is a CPU operation that can happen entirely in your browser. There is no technical need for a server, an account, or even an internet connection once the page has loaded.

Tools that require signup do so for business reasons: to collect your email for marketing, to enforce usage tiers that push you toward paid plans, and to track your usage patterns. None of this is necessary for the core functionality of compressing an image.

When evaluating image compression tools, consider the friction-to-value ratio. The best tools deliver value with the least friction. Browser-based tools inherently win on every friction dimension because the processing is free for the tool provider.

This also applies to GDPR and privacy regulations. Under the GDPR, uploading images to a server constitutes data processing and requires legal basis, a privacy policy, and potentially a data processing agreement. Browser-based tools that never transmit your files sidestep this entirely.

The weight of images on the modern web.

According to the HTTP Archive's 2025 Web Almanac, the median web page weighs approximately 2.5 megabytes on desktop and 2.2 megabytes on mobile. Images account for roughly 50 percent of that total page weight, making them the single largest component of most web pages.

28 percent of image bytes could be saved by serving images at their actual display dimensions. An additional 20 percent could be saved by converting to modern formats like WebP. The average page includes 30 image requests, many of which are uncompressed or minimally compressed.

This is why batch compression matters. If you are managing a website with hundreds of images, compressing them one at a time is impractical. A batch tool that processes 20 images in a single operation makes the optimization workflow viable.

How to batch compress images in your browser.

Step 1: Open SammaPix Compress in your browser. No account, no download, no installation.

Step 2: Drag and drop up to 20 images at once. Supports JPEG, PNG, WebP, GIF, AVIF, and HEIC.

Step 3: Set your quality level. 80 is the sweet spot for web use. It delivers 50 to 80 percent file size reduction with no visible quality difference.

Step 4: Watch the compression happen in real time. Each file shows its original size, compressed size, and percentage saved.

Step 5: Download each image individually or all at once as a ZIP on Pro.

The entire process happens in your browser. If you open your browser's network inspector during compression, you will see zero outbound requests.

When server-side compression is the better choice.

Being honest about trade-offs: there are cases where server-side compression genuinely makes more sense.

Automated pipelines. If you need to compress thousands of images automatically as part of a build process or CMS workflow, an API-based service like TinyPNG's API, Cloudinary, or imgix is the right tool. Browser-based tools require manual interaction.

Maximum compression with MozJPEG or AVIF. Server-side tools can use advanced codecs like MozJPEG for 5 to 10 percent better JPEG compression, or AVIF encoding which is computationally expensive. For the absolute maximum compression at the highest quality, server tools have a small edge. For 95 percent of use cases, the browser result is indistinguishable.

WordPress and CMS plugins. If you run a WordPress site, plugins like ShortPixel or Imagify automatically compress images when you upload them. This is convenient because it requires zero manual effort.

Combining compression with other image operations.

Compression alone is rarely enough. The optimal image workflow for web content involves multiple steps.

Resize first, then compress. A 4000 pixel image resized to 1200 pixels before compression saves significantly more than compression alone.

Convert to WebP. WebP is 25 to 34 percent smaller than JPEG at equivalent visual quality, with 97 percent browser support.

Strip EXIF metadata. EXIF data like GPS coordinates, camera model, and timestamps adds 10 to 100 kilobytes per image and is a privacy risk.

Rename for SEO. Files named IMG underscore 4521 dot jpg contribute nothing to search rankings. AI-generated descriptive names improve image SEO.

The combined effect of resize plus format conversion plus compression can reduce a batch of images from 100 megabytes to under 5 megabytes with no visible quality loss.

Batch compression on mobile devices.

One of the biggest advantages of browser-based compression is that it works on any device with a modern browser, including phones and tablets. There is no app to install. You open the tool in Safari or Chrome, select your photos, and they are compressed locally.

This is particularly useful for iPhone users. iPhones shoot in HEIC format by default, which is not universally supported. You can convert HEIC to JPEG or WebP, then compress the result, all on your phone, all in the browser.

Performance on mobile is solid. A modern iPhone or mid-range Android phone with 6 gigabytes of RAM can compress a batch of 20 images in 10 to 20 seconds. Even on older devices, the process completes within a minute for a full batch.

Originally published at sammapix.com

Try it free: SammaPix — 27 browser-based image tools. Compress, resize, convert, remove background, and more. Everything runs in your browser, nothing uploaded.

How AI Image Renaming Boosts Your SEO (2026 Guide)

Luca Sammarco — Mon, 13 Apr 2026 14:02:17 +0000

Why image filenames matter for SEO

Google uses image filenames as one of several signals to understand what an image depicts. This is documented directly in Google's official image guidelines : "The filename can give Google clues about the subject matter of the image. For example, my-new-black-kitten.jpg is better than IMG00023.JPG." That is a direct quote from the documentation.

Google Image Search drives a significant volume of traffic for content publishers, e-commerce sites, travel bloggers, food websites, and product marketers. According to research aggregated by Moz , Google Images represents approximately 22.6% of all web searches - making it the second-largest search surface after Google's main index. Optimizing for it is not optional if you are serious about organic visibility.

The filename is processed before the page is fully crawled. When Googlebot discovers an image URL such as /images/golden-gate-bridge-sunset.jpg , it extracts the filename tokens immediately. This gives descriptive filenames an early relevance advantage before alt text, captions, or surrounding content are even evaluated.

The problem: camera-generated filenames tell Google nothing

Every camera, smartphone, and screenshot tool generates filenames automatically. They are designed for the camera's internal file management- not for search engines or human readers. The result is a collection of filenames that convey zero semantic information:

IMG_0001.jpg, IMG_0002.jpg- iPhone sequential numbering
DSC_1234.CR2, _MG_5678.NEF- DSLR raw files
Screenshot 2026-02-14 at 09.32.11.png- macOS screenshots
20260214_093211.jpg- Android timestamp filenames
image (1).png, image (2).png- browser download defaults

When Google encounters these filenames, it receives no useful signal. The image is essentially anonymous to the search engine until it cross-references alt text, surrounding text, and its own computer vision analysis. You are leaving ranking potential on the table with every unnamed image.

The traditional solution is to rename each image manually before uploading. For a photographer or blogger with a library of hundreds or thousands of images, that is an enormous time investment. Manually generating SEO-appropriate filenames for 300 product photos is realistically a full workday. This is the exact problem AI image rename tools are built to solve.

How AI understands image content and generates descriptive filenames

Modern AI vision models can analyze an image and identify its content with remarkable accuracy- subjects, scenes, objects, colors, spatial relationships, and contextual details. This is the same category of technology that powers Google Lens, Apple's Visual Look Up, and Pinterest's visual search.

When you drop an image into an AI rename tool, the process works as follows. First, a thumbnail of the image is generated client-side to reduce the data payload. This thumbnail is then sent to a vision model- in SammaPix's case, Google Gemini Flash- which analyzes the visual content. The model returns a structured description of what it sees. The tool then applies SEO naming rules to that description: lowercase only, words separated by hyphens, relevant primary subject first, location or color modifiers second, no stop words or filler.

The result is a filename that would take a human several seconds to write per image, generated in under two seconds per image at scale. For a batch of 50 product photos, an AI rename tool completes in roughly a minute what would otherwise require 30–45 minutes of manual effort.

Before and after: what AI image renaming actually produces

The transformation is straightforward to illustrate. Here are real-world examples across different image categories:

Before (camera default) |
After (AI rename) |

IMG_0023.jpg |
golden-gate-bridge-sunset-san-francisco.jpg |

DSC_4891.jpg |
espresso-coffee-latte-art-white-ceramic-cup.jpg |

image (14).png |
minimal-dark-mode-dashboard-ui-analytics.png |

20260201_174302.jpg |
woman-hiking-mountain-trail-snow-winter-alps.jpg |

_MG_0112.CR2 (converted) |
red-maple-leaf-autumn-forest-ground-macro.jpg |

Notice the pattern in the AI-generated filenames. The primary subject appears first (golden-gate-bridge, espresso-coffee, woman-hiking). Descriptive modifiers follow (sunset, latte-art, mountain-trail). Location or context comes last where relevant (san-francisco, alps). Every word is a potential keyword match.

How SammaPix AI Rename works

SammaPix AI Rename is built on Google Gemini Flash, a multimodal vision model optimized for speed without sacrificing accuracy on visual understanding tasks. The workflow is entirely private by design: your original images never leave your browser. Only a compressed thumbnail is sent to the API for analysis.

The technical pipeline works in three steps. First, the tool generates a thumbnail of your image locally using the browser Canvas API- typically at 512 pixels on the longest side, which is more than sufficient for content recognition while keeping the API payload small. Second, the thumbnail is sent to Gemini Flash with a structured prompt that instructs the model to identify the subject, scene, dominant colors, and any distinguishing details, then format the output as a valid SEO filename. Third, the generated filename is validated against SEO rules (lowercase, hyphens only, no special characters, capped at a reasonable length) before being presented to you.

You can rename a single image or drop an entire batch of files. The tool processes them in parallel and presents a preview of each suggested filename before you download. You can edit any suggestion manually if you want to adjust the output- the AI gives you a strong starting point, but you always have final control. Download your renamed images individually or as a ZIP archive.

SEO best practices for image filenames

Whether you rename manually or use an AI tool, the rules for SEO-optimized image filenames are consistent and well-established. Following them ensures your images have the best possible chance of appearing in Google Image Search results.

Use hyphens, not underscores or spaces

Google treats hyphens as word separators. A filename of red-ceramic-coffee-mug.jpg is parsed as four separate words: red, ceramic, coffee, mug. Underscores are treated as connectors - Google reads red_ceramic_coffee_mug.jpg as a single compound token. Spaces in filenames are encoded as %20 in URLs, which creates messy links and can cause crawl issues. Always use hyphens.

Keep filenames lowercase

URLs are case-sensitive on most server configurations. A file named Golden-Gate-Bridge.jpg and golden-gate-bridge.jpg can be treated as two different resources by the server, creating duplicate content issues. Lowercase filenames also look cleaner in image URLs and match the convention used by all major web platforms.

Be descriptive and specific

Generic filenames like photo.jpg or image1.jpg provide no ranking benefit. Aim for 3 to 5 descriptive keywords that accurately reflect the image content. Include the primary subject, any relevant modifiers (color, material, style), and location or context where meaningful. Avoid keyword stuffing- a filename like buy-cheap-coffee-mug-online-store-sale.jpg reads as spam and provides no additional ranking benefit over a clean, accurate descriptive name.

Drop stop words and filler

Words like "a", "the", "of", "and", "with" add length to the filename without contributing to keyword relevance. Keep filenames tight. Prefer wooden-dining-table-modern-interior.jpg over a-photo-of-a-wooden-dining-table-in-a-modern-interior.jpg .

Match the filename to the page topic

A travel blog post about Tokyo should feature images with filenames that include relevant location keywords. An e-commerce product page for running shoes should have filenames reflecting the specific product model, color, and use case. Contextual alignment between the filename, alt text, page title, and surrounding content reinforces your topical relevance signal to Google.

Alt text vs filename vs title attribute: understanding the difference

These three image attributes serve different purposes and carry different weight as SEO signals. Confusing them leads to missed optimization opportunities. Here is how each one functions:

Image filename (the URL path)

The filename is part of the image's URL. It is processed by Google before the page HTML is fully parsed. It signals topical relevance early in the crawl cycle. It cannot be changed without also updating every reference to the image URL on your site. This is why getting filenames right before upload matters- changing them after indexing requires redirects.

Alt text (the alt attribute)

Alt text is the most important on-page image SEO signal. It lives in the HTML as . It serves two critical functions: it describes the image to Google's crawler in the context of the page, and it provides text to screen readers for accessibility. Alt text should describe the image as it relates to the page content- not as a generic description and not as a keyword dump.

Title attribute

The title attribute on an image appears as a tooltip when a user hovers over it. It is the weakest of the three signals for SEO purposes- Google gives it minimal weight. Focus your optimization effort on filename and alt text first. The title attribute is useful for user experience (hover tooltips) but should not be treated as a primary SEO lever. According to web.dev's performance guidelines , investing time in meaningful alt text and optimized filenames returns far more value than filling in title attributes.

Signal |

SEO weight |

Purpose |

Required? |

Filename |

High |

Early relevance signal at crawl time |
Yes (always present) |

Alt text |

Very high |

Contextual relevance + accessibility |
Yes (required for accessibility) |

Title attribute |

Low |

Tooltip on hover, minimal SEO value |
Optional |

Step-by-step: rename images for SEO with SammaPix AI Rename

Using the SammaPix AI Rename tool takes under five minutes for a batch of 30 to 50 images. Here is the complete workflow:

Step 1 - Sign in (free account required)

AI Rename requires a free account to prevent API abuse. Click "Sign in with Google" or "Sign in with GitHub"- no forms, no passwords, no email verification. The entire sign-in flow takes under 15 seconds.

Step 2 - Drop your images into the tool

Drag and drop a batch of images onto the drop zone, or click to open a file picker. You can mix JPEG, PNG, WebP, and HEIC files in the same batch. The tool accepts up to 5 images per session on the free plan and up to 200 on Pro.

Step 3 - Review AI-generated suggestions

For each image, the tool displays the original filename alongside the AI-generated suggestion. Review each suggestion- in our testing, Gemini Flash is accurate on about 95% of images without any manual correction needed. For the edge cases (abstract art, macro photography with ambiguous subjects), you can click the filename field and edit it directly.

Step 4 - Download your renamed images

Download individual images by clicking the download button on each card, or click "Download All as ZIP" to get the entire batch in a single archive. The original files are unchanged- the tool creates new copies with the new filenames. Your files never left your device except for the thumbnail sent to Gemini for analysis.

Step 5 - Upload and update alt text

Upload the renamed images to your CMS, e-commerce platform, or image hosting service. Take the opportunity to update the alt text for each image to match the new descriptive filename- use the filename as a starting point, then rephrase it as a natural sentence. For example, golden-gate-bridge-sunset-san-francisco.jpg becomes alt="Golden Gate Bridge at sunset viewed from San Francisco" .

If you also need to reduce file sizes before uploading- which is almost always a good idea for web performance- run your images through the SammaPix Compress tool or Convert to WebP first, then rename. The order does not affect the SEO benefit, but compressing before renaming means you only keep one optimized copy of each file. For more on compression best practices, see our guide on compressing images without losing quality .

Frequently asked questions

Does image filename actually affect SEO rankings?

Yes. Google's own documentation confirms that image filenames are a relevance signal for Google Images. While filename alone is not a ranking factor for your main page results, it directly affects discoverability in Google Image Search- which is responsible for approximately one in five all web searches. For content-heavy sites, photography portfolios, food blogs, and e-commerce stores, optimizing image filenames is one of the highest-return technical SEO tasks available.

Should I use hyphens or underscores in image filenames for SEO?

Always use hyphens. Google treats hyphens as word separators- golden-gate-bridge is parsed as three separate keywords. Underscores are treated as connectors- golden_gate_bridge is treated as one compound token. Hyphens give you better keyword matching across all three words.

What is the difference between alt text and image filename for SEO?

The filename is part of the image URL and gives Google an initial relevance signal before the full page is crawled. The alt text lives in the HTML and provides contextual relevance within the page- it also serves as accessibility text for screen readers. Both are important. A descriptive filename without good alt text leaves half the signal unused, and vice versa. The two work together to reinforce topical relevance.

Is it worth renaming old images already indexed by Google?

Yes, but you must set up proper 301 redirects from the old image URL to the new one. If you rename and upload without redirects, any existing backlinks or indexed image URLs will return 404 errors, causing you to lose existing ranking equity. If your images are not yet indexed or do not have inbound links, you can rename and re-upload without redirects.

How many keywords should I include in an image filename?

Aim for three to five descriptive keywords. Fewer than three often leaves relevant context out. More than five starts to look like keyword stuffing to Google's algorithms and produces unwieldy filenames. The sweet spot is a filename that reads naturally and accurately describes the image- if it sounds like an honest description of what you see in the image, it is probably correct.

Originally published at sammapix.com

Try it free: SammaPix — 27 browser-based image tools. Compress, resize, convert, remove background, and more. Everything runs in your browser, nothing uploaded.

Methodology

Luca Sammarco — Mon, 13 Apr 2026 14:02:06 +0000

Alt text is no longer optional. It is a ranking factor, an accessibility requirement, and increasingly generated by AI. But how accurate are these AI-generated descriptions really? I tested three leading models: Google Gemini 2.5 Flash, GPT-4o, and Claude 3.5 Sonnet. I ran all three on 200 real photographs across 5 categories, and honestly, the results surprised me. Some models consistently misidentified objects, others generated descriptions too generic for SEO value, and one model stood out for e-commerce product photos. This is the first public benchmark comparing AI alt text quality with actual accuracy scores, SEO usefulness ratings, and accessibility compliance checks. Every image was scored on 4 criteria: factual accuracy, SEO keyword inclusion, accessibility usefulness, and appropriate length.

Methodology

I selected 200 photographs split evenly across five categories: portraits (40), landscapes (40), e-commerce products (40), screenshots/UI (40), and food (40). Images were sourced from real production environments, including my own travel photography, client e-commerce catalogs, open-source UI projects, and stock photo libraries.

Each image was processed through all three models using their respective APIs: Google Gemini 2.5 Flash via the Gemini API, GPT-4o via OpenAI's vision endpoint, and Claude 3.5 Sonnet via Anthropic's messages API. Each model received the same prompt: "Generate alt text for this image. The alt text should be concise, descriptive, and suitable for SEO and screen readers."

I scored every output on four criteria, each rated 1 to 10. Factual Accuracy: Does it correctly describe what is in the image? SEO Value: Does it include relevant keywords a real user would search for? Accessibility: Would a screen reader user understand the image? Length: Is it the right length, where under 10 words is too vague and over 40 words creates clutter?

The overall score is the unweighted average of all four criteria. I scored every single output manually, not with another AI model. I wanted to make sure there was real human judgment on factual accuracy and real-world usefulness.

Overall results

Here are the aggregate scores across all 200 images. Gemini 2.5 Flash scored 8.2 on accuracy, 7.8 on SEO value, 7.5 on accessibility, 8.0 on length, for an overall score of 7.9. GPT-4o scored 8.5 on accuracy, 7.2 on SEO value, 8.1 on accessibility, 7.3 on length, for an overall score of 7.8. Claude 3.5 Sonnet scored 8.7 on accuracy, 6.9 on SEO value, 8.4 on accessibility, 6.8 on length, for an overall score of 7.7.

I was surprised by how close the overall scores are. Just 0.2 points separate first from last. But the individual criteria tell a very different story. Claude is the most accurate model but scores lowest overall because its descriptions are consistently too long. Gemini wins not because it is the smartest, but because it produces the most practical alt text: the right length, with the right keywords, at the right level of detail.

Results by category

The aggregate scores hide significant differences across image types. For portraits, Claude wins with an accuracy score of 8.9, the highest single-category score in the entire benchmark. Claude excels at detecting emotions, context clues, and even approximate age ranges. The tradeoff is length: Claude averaged 48 words for portraits, which is excessive for alt text.

For landscapes, Gemini wins with the highest overall category score of 8.2. What sets Gemini apart is its ability to identify specific locations. Where GPT-4o might describe "a mountain range with a lake in the foreground," Gemini consistently identified landmarks like "Mount Fuji reflected in Lake Kawaguchi at sunrise."

For e-commerce products, Gemini dominates with an SEO score of 8.4, the highest individual SEO score in the entire benchmark. Gemini naturally includes product-relevant keywords that match actual search queries: material, color, product type, and style descriptors.

For screenshots and UI images, GPT-4o dominates with a category-best accuracy of 8.8. GPT-4o's strength is its ability to read text embedded in images, including button labels, menu items, error messages, and code snippets.

For food, it is a virtual tie between GPT-4o and Gemini at 8.0 versus 7.9 overall. Both models are strong at identifying ingredients and dish types.

5 key findings

Finding one: Gemini generates the most SEO-friendly descriptions. Gemini 2.5 Flash scored 7.8 out of 10 on SEO value, the highest of any model. For product images, Gemini included brand names, materials, and colors 87% of the time.

Finding two: Claude is the most accurate but often too verbose. Claude 3.5 Sonnet achieved an 8.7 accuracy score, 0.5 points above Gemini. However, Claude averaged 45 words per description compared to Gemini's 22 words.

Finding three: All three models fail on culturally-specific content. When I tested images of traditional clothing, religious ceremonies, and regional food, all three models showed significant blind spots. Across the full test set, 31% of culturally-specific items were misidentified or described too generically.

Finding four: GPT-4o is the best model for screenshots and UI images. GPT-4o scored 8.8 on accuracy for screenshots, the highest single-model, single-category accuracy score. Its advantage is OCR: GPT-4o reads and incorporates text visible in the image.

Finding five: For e-commerce, AI alt text outperforms human-written alt text 73% of the time. I compared AI-generated alt text to existing human-written alt text for 40 e-commerce product images. The reason is predictable: humans tend to write alt text that is either too short or stuffed with marketing language. AI models produce descriptive, natural-language alt text that better matches how users actually search.

Which model should you use?

For e-commerce product images, use Gemini 2.5 Flash. Highest SEO value, optimal length, and the fastest and cheapest per image.

For blog and editorial content, use GPT-4o. Best balance of accuracy, SEO value, and readability, averaging 26 words.

For accessibility compliance, use Claude 3.5 Sonnet. Highest accessibility score and factual accuracy, though you may want to trim descriptions to 30 words or fewer.

For batch processing at scale, use Gemini 2.5 Flash. Fastest response time at 0.8 seconds per image versus 1.4 seconds for GPT-4o and 1.9 seconds for Claude.

How SammaPix uses AI alt text

SammaPix uses Gemini 2.5 Flash for its AI Alt Text generator, based on the results of this benchmark. The choice was driven by three factors: highest overall score of 7.9, best SEO value of 7.8, and optimal length averaging 22 words. The tool is browser-based, your images are processed locally, and the free tier includes 10 images per day with no account required.

Originally published at sammapix.com

Try it free: SammaPix — 27 browser-based image tools. Compress, resize, convert, remove background, and more. Everything runs in your browser, nothing uploaded.

How to Remove EXIF Data and Protect Your Privacy

Luca Sammarco — Mon, 13 Apr 2026 09:23:46 +0000

What is EXIF data?

EXIF stands for Exchangeable Image File Format. It is a standard for embedding technical metadata directly inside digital image files- specifically JPEG, TIFF, and RAW formats. The standard was developed by the Japan Electronic Industries Development Association in 1995 and is now used by virtually every digital camera and smartphone camera app in the world.

When you press the shutter, the camera writes a block of data into the file alongside the image pixels. This block is invisible when you view the photo but can be read by any software that knows where to look- including free online tools, desktop apps, and command-line utilities. The full technical specification is documented on the EXIF Wikipedia page.

{title}

{detail}

Why EXIF data is a real privacy risk

The risks are not hypothetical. There are documented, high-profile cases where EXIF metadata led directly to serious privacy breaches.

Home address from a Marketplace listing

Selling something on Craigslist, Facebook Marketplace, or Vinted? If you photograph the item at home and upload the original file, the GPS coordinates in the EXIF data reveal your home address to every buyer- and everyone else who downloads the image. Stalking cases have been traced back to exactly this scenario.

The John McAfee case (2012)

Vice Media published an exclusive interview with John McAfee while he was in hiding in Guatemala. The accompanying iPhone photos had GPS coordinates embedded in their EXIF data. The coordinates were publicly readable in the published images, revealing McAfee's exact location to authorities. He was found and detained shortly after. The error- leaving GPS metadata intact in published photos- was entirely avoidable.

Military operational security

In 2007, US Army soldiers photographed newly delivered Apache helicopters at a base in Iraq and posted the images to the internet. The GPS metadata embedded in the photos revealed the exact coordinates of the military base. The incident prompted the US Army to update its digital photography and social media policy. The same risk applies to anyone working in a sensitive location.

Routine and pattern exposure

A series of photos posted to a public social account over time can map an entire daily routine. GPS timestamps reveal where you work, where your children attend school, and what routes you travel regularly. This kind of pattern data is exactly what stalkers and abusive ex-partners look for- and it is sitting in plain sight for anyone who knows to look at EXIF metadata.

Every smartphone photo can carry precise GPS coordinates in its metadata - Photo by Jonas Leupe on Unsplash

Do social platforms strip EXIF automatically?

Instagram, Facebook, and X (Twitter) do strip most EXIF metadata when you upload a photo through their apps. This sounds reassuring- but it is not a reliable privacy strategy.

{item}

The only reliable approach is to strip EXIF data from the source file before you share it anywhere- so the metadata never reaches any platform in the first place.

How to remove EXIF data using SammaPix EXIF Viewer

SammaPix EXIF Viewer processes your photos entirely in the browser- no upload, no server, no account required. Here is the complete process.

{title}

{body}

Free tool- no upload, no signup

Strip EXIF data from your photos now - SammaPix EXIF Viewer

EXIF privacy tips for social media sharing

Even if you trust the platforms you share to, building good habits around EXIF data reduces your overall digital footprint. Here are the most important practices.

Always strip GPS before listing anything for sale

Marketplace listings (Facebook, Craigslist, Depop, eBay) are the highest-risk context. The images are often downloadable by anyone, and sellers typically photograph items in their home. Remove GPS data before every listing photo you upload.

Be careful with iMessage and email

WhatsApp strips EXIF before sending. iMessage does not- photos sent via iMessage retain all original metadata including GPS. Email attachments also preserve metadata. If you are sending photos that include sensitive GPS data (your home, a private event, a confidential location), strip them first.

Disable GPS for your camera app if you do not need it

On iPhone: Settings -> Privacy & Security -> Location Services -> Camera -> Never. On Android: Camera settings -> Location tag -> Off. Photos taken without GPS enabled will never have coordinates to worry about. The downside is losing location data for travel photography- so it is a tradeoff worth considering per use case.

Use a dedicated workflow for client photos

If you are a photographer delivering images to clients, EXIF data contains your camera serial number, lens details, and shooting settings. Some clients or agencies specify that delivered files should have metadata stripped- particularly in line with IPTC photo metadata standards. Build EXIF removal into your export workflow rather than doing it manually per job.

Compress and strip in one step

If you are also optimizing images for web use, SammaPix Compress strips all EXIF metadata automatically as part of the compression process. You get a lighter file with no metadata in a single operation- no separate EXIF removal step needed.

Building good metadata hygiene into your photo sharing workflow protects your privacy - Photo by John Schnobrich on Unsplash

Alternative methods for removing EXIF data

Windows: built-in File Properties

Right-click any image file -> Properties -> Details tab -> "Remove Properties and Personal Information." You can strip all metadata or specific fields. Works on Windows 10 and 11 with no software to install. Limitation: one file at a time, no batch processing.

macOS: Photos app export

In the macOS Photos app, select your images, go to File -> Export -> Export Photos, and uncheck "Location Information." This exports copies without GPS data. Note: macOS Preview does not strip EXIF on export, so the Photos app method is the native approach for GPS removal on Mac.

iOS 17+: Share sheet

In iOS 17 and later, when you share a photo via the Share sheet, tap "Options" at the top and you will find a "Location" toggle. Disabling it strips GPS coordinates from the shared copy. This is the fastest mobile method but only works on iOS 17+ and only during the share action.

Command line: ExifTool

ExifTool by Phil Harvey is the authoritative open-source tool for batch EXIF processing. To remove all GPS fields: exiftool -gps:all= -overwrite_original photo.jpg. Use the -r flag for recursive directory processing. Powerful for automation but requires terminal comfort and installation.

FAQ

{q}

{a}

Originally published at sammapix.com

Try it free: SammaPix — 27 browser-based image tools. Compress, resize, convert, remove background, and more. Everything runs in your browser, nothing uploaded.

HEIC to WebP Converter: Why You Should Skip JPG in 2026

Luca Sammarco — Mon, 13 Apr 2026 09:23:35 +0000

Why every iPhone shoots HEIC by default.

Apple introduced HEIC, or High Efficiency Image Container, as the default photo format with iOS 11 and the iPhone 7 in 2017. HEIC uses the HEVC codec for image compression, which is significantly more efficient than JPEG's compression from 1992.

The practical benefit is substantial: HEIC files are roughly 50% smaller than equivalent JPEG files at the same visual quality. For a phone that stores thousands of photos, this means approximately half the storage space consumed. Apple made this switch specifically to manage storage on devices with fixed capacity.

As of 2025, there are approximately 1.46 billion active iPhones worldwide according to Statista. Every single one of them is producing HEIC photos by default. This makes HEIC one of the most widely produced image formats in the world, even though most websites and platforms still cannot display it natively.

If you have ever tried to upload an iPhone photo to a website and received an unsupported format error, HEIC is the reason. The format is excellent for storage but needs to be converted before most web platforms can use it. The question is: what should you convert it to? If you have been defaulting to JPG, you have been making the wrong choice.

Why converting HEIC to JPG is a mistake.

The most common workflow for using iPhone photos on the web is: take photo in HEIC, convert to JPG, then upload. This is the wrong approach.

HEIC is a lossy format. When the iPhone captures a photo, it applies HEVC compression and discards some image data permanently. The resulting HEIC file is already compressed. When you convert that HEIC to JPG, the converter decodes the HEIC data and re-encodes it using JPEG compression. This re-encoding applies a second round of lossy compression, discarding even more data.

This is called generation loss. Each time you re-encode lossy data into another lossy format, quality degrades. The degradation is especially visible around high-contrast edges, text, and fine details. And because JPEG's compression is less efficient than HEIC's, the resulting JPG file is often larger than the original HEIC despite being lower quality.

Think of it like photocopying a photocopy. The original is sharp, but each copy of a copy gets progressively blurrier. HEIC to JPG is exactly this: you are making a lossy copy of a lossy original, using an older and less efficient copying method. The result is predictably worse on both dimensions, larger file size and lower quality.

HEIC to WebP: the advantages.

WebP uses the VP8 codec for lossy and VP8L codec for lossless compression, which are modern compression algorithms comparable in efficiency to HEVC. Converting HEIC to WebP still involves re-encoding, but the quality preservation is significantly better than converting to JPG.

Better compression efficiency means WebP produces smaller files than JPG at equivalent visual quality, so you can use a higher quality setting and still get a smaller file. Modern artifact handling means WebP's compression produces less visible artifacting than JPEG, particularly around edges and in areas with fine detail. Alpha channel support means that unlike JPG, WebP supports transparency. Animation support means WebP can replace GIF as well. WebP is a web-native format designed by Google specifically for web use, so it integrates perfectly with Core Web Vitals optimization.

The bottom line: HEIC and WebP are both modern formats that speak the same language of efficient compression. Converting between them is like translating between two closely related languages, very little is lost. Converting HEIC to JPG is like translating a modern novel into an ancient language with a limited alphabet. Something is inevitably lost.

File size comparison: HEIC vs JPG vs WebP.

For a typical iPhone 15 Pro photo at 4032 by 3024 pixels, 12 megapixels, converted at quality 80: The original HEIC is about 1.8MB. Converting to JPG produces a 2.4MB file, 33% larger than the original, with visible generation loss. Converting to WebP produces a 1.6MB file, 11% smaller than the original, with minimal quality loss. Converting to PNG for lossless preservation produces a 12MB or larger file.

The key insight: converting HEIC to JPG produces a file that is both larger and lower quality than the original. Converting to WebP produces a file that is smaller than the HEIC original while preserving nearly all the visual quality.

These numbers are based on typical photographic content. Images with lots of fine detail like text, foliage, and hair show even larger gaps between JPG and WebP quality because JPEG's block-based compression creates more visible artifacts in high-frequency areas.

WebP browser support in 2026.

One of the historical arguments for converting to JPG was browser compatibility. WebP was introduced by Google in 2010, but Safari did not add support until Safari 14 in September 2020. This made WebP unreliable for a significant portion of web traffic.

In 2026, that argument is obsolete. According to Can I Use data, WebP has over 97% global browser support. Every current version of Chrome, Firefox, Safari, Edge, and Opera supports WebP natively. The only browsers lacking support are Internet Explorer 11 and very old mobile browsers with negligible market share.

For the remaining 3% of browsers, you can serve a JPG fallback using the HTML picture element. But for most sites, serving WebP directly without a fallback is perfectly acceptable. Google, Facebook, Netflix, and essentially every major website now serves WebP by default.

HEIC on social media and messaging apps.

If you upload HEIC photos directly to social platforms, something happens behind the scenes that most people do not realize: the platform converts your HEIC to JPEG automatically, using its own compression settings. You have no control over the quality.

Instagram. Instagram accepts HEIC uploads but immediately converts them to JPEG at its own quality level. The platform also resizes images to 1080 pixels on the longest side. If you upload a 12 megapixel HEIC, Instagram does three things at once: decode HEIC, resize from 4032 pixels to 1080 pixels, and re-encode as JPEG. You get a triple quality hit. Converting to WebP first and resizing to 1080 pixels yourself gives you control over the compression quality.

WhatsApp. WhatsApp compresses every image to roughly 70 to 100 kilobytes and resizes to approximately 1600 pixels regardless of the original format or size. Sending a HEIC file through WhatsApp means the app first converts HEIC to JPEG, then compresses that JPEG aggressively. The result is significantly degraded.

Facebook and Twitter/X. Both platforms accept HEIC and convert internally. Facebook re-encodes to JPEG at moderate quality. Twitter does the same but slightly more aggressively. In both cases, converting to a well-optimized WebP or JPEG yourself before uploading produces noticeably better results because you control the compression quality rather than letting the platform's algorithm decide.

The pattern is consistent across all platforms: if you let the platform handle conversion, you get their lowest-common-denominator compression. If you convert beforehand, you get better quality because the platform has less work to do.

HEIC vs WebP vs AVIF: which modern format to choose.

WebP is not the only modern format. AVIF, or AV1 Image File Format, is even newer, released by the Alliance for Open Media in 2019. It offers superior compression to both WebP and HEIC.

HEIC has about 18% browser support through Safari only. WebP has over 97% browser support. AVIF has about 93% browser support. In terms of compression efficiency, HEIC is excellent, WebP is very good, and AVIF is the best. For encoding speed, HEIC and WebP are fast while AVIF is 5 to 10 times slower. All three support transparency. HEIC has licensing fees while WebP and AVIF are royalty-free. HEIC is best for iPhone storage, WebP is the safest choice for web images, and AVIF offers the best quality-to-size ratio for web images.

AVIF produces files 20 to 30% smaller than WebP at the same visual quality. However, it has two significant drawbacks today: encoding is 5 to 10 times slower than WebP which matters for batch processing, and browser support is at about 93% versus WebP's 97% or higher. The gap is closing, but for production websites that need universal compatibility, WebP remains the safer choice.

The practical recommendation: convert HEIC to WebP for web use today. If you can serve AVIF with a WebP fallback using the picture element, that gives you the best of both worlds. But WebP alone is an excellent choice that works everywhere.

When JPG is still the right choice.

Despite WebP's clear advantages, there are a few scenarios where JPG remains the better target format. Email attachments to clients using older software, printing services that only accept JPG, PNG, or TIFF, legacy CMS platforms that do not handle WebP uploads, and older versions of Photoshop that require a plugin for WebP.

For all web-facing use cases in 2026, WebP is the correct choice over JPG.

The optimal conversion workflow.

For web use including blogs, websites, and e-commerce: convert HEIC to WebP at quality 80, resize to display dimensions typically 1200 to 1920 pixels wide, strip EXIF data for privacy, compress further if needed, and upload the optimized WebP file.

For editing in Photoshop or Lightroom: convert HEIC to PNG lossless to preserve maximum data for editing, edit in your application, then export the final result as WebP for web or JPG for print.

For archival storage: keep the original HEIC files since they are already efficiently compressed and contain all the metadata. Only convert when you need to use the photo somewhere that does not support HEIC.

Batch converting HEIC photos from your iPhone.

If you have dozens or hundreds of HEIC photos from a trip, event, or product shoot that need to be converted for web use, doing them one at a time is not practical. A batch workflow is essential.

The batch conversion process. First, transfer HEIC photos to your computer via AirDrop, iCloud, or USB cable. AirDrop is fastest for small batches but for large libraries USB transfer avoids compression. Second, open the HEIC Converter in your browser and select all HEIC files at once. SammaPix processes up to 20 images simultaneously on the free plan and 500 on Pro. Third, set output format to WebP and quality to 80. All files are converted with the same settings. Fourth, download all converted files. On Pro you can download everything as a single ZIP file.

The entire process runs in your browser. No files are uploaded to any server, which makes it fast even for large batches. The speed depends on your device's processor, not your internet connection. A modern MacBook can convert 50 HEIC photos to WebP in under 30 seconds.

Privacy: removing location data during conversion.

Every HEIC photo taken on an iPhone contains embedded EXIF metadata. This includes information most people do not realize is there. GPS coordinates showing the exact latitude and longitude where the photo was taken, accurate to within a few meters. Timestamp with the exact date and time including timezone. Device information including iPhone model, iOS version, lens used, and focal length. Camera settings like aperture, shutter speed, ISO, and exposure compensation. Unique identifiers including lens make, software version, and sometimes a unique image ID.

When you convert HEIC to another format, some converters preserve this metadata and some strip it. If you are publishing photos on a website or sharing them publicly, you should always strip EXIF data, especially GPS coordinates. Publishing a photo with GPS data reveals your exact location, your home address, your workplace, or the locations you frequent.

The recommended approach is to include EXIF removal as a step in your conversion workflow. After converting HEIC to WebP, run the files through the EXIF Remover to strip all metadata.

If you are managing images for a website and want to automate the naming as well, the AI Rename tool can generate SEO-friendly filenames based on image content, and the Alt Text Generator creates accessibility-compliant alt descriptions. Combined with EXIF removal and WebP conversion, this gives you a complete image optimization pipeline.

Frequently Asked Questions.

Why is HEIC to WebP better than HEIC to JPG?

Both HEIC and WebP use modern compression algorithms, while JPG uses technology from 1992. Converting HEIC to WebP preserves more visual quality because both formats handle compression similarly. Converting HEIC to JPG introduces a generation loss from re-encoding into an older, less efficient format. WebP files are also 25 to 34% smaller than equivalent JPG files at the same visual quality.

Does every iPhone shoot in HEIC format?

Every iPhone since the iPhone 7 running iOS 11 released in 2017 shoots in HEIC by default. You can change this in Settings, Camera, Formats, switching from High Efficiency to Most Compatible. However, HEIC produces better quality at smaller file sizes, so keeping HEIC and converting to WebP for web use is the optimal workflow.

Can all browsers display WebP images in 2026?

WebP has over 97% global browser support as of 2026 according to Can I Use data. Every major browser including Chrome, Firefox, Safari, Edge, and Opera supports WebP. The only browsers that lack support are very old versions like Internet Explorer 11, which has less than 0.3% global usage.

When should I still convert HEIC to JPG instead of WebP?

JPG is still the better choice when sending email attachments to clients with very old software, uploading to printing services that only accept JPG, working with legacy CMS platforms that do not support WebP, and sharing files with older versions of Photoshop. For all web use cases, WebP is superior.

Can I convert HEIC to WebP on my iPhone?

Yes. You can use a browser-based converter like SammaPix directly in Safari or Chrome on your iPhone. Open the HEIC Converter tool, select your photos from the camera roll, and convert them to WebP without installing any app. The conversion runs entirely on your device and your photos are never uploaded to a server. This works on any iPhone running iOS 14 or later.

Does converting HEIC to WebP remove EXIF data?

It depends on the tool. Some converters strip EXIF metadata during conversion, while others preserve it. HEIC photos from iPhones contain GPS coordinates, camera settings, and timestamps. If privacy is a concern, use a dedicated EXIF remover after converting to WebP, or use a workflow that includes EXIF stripping as a step. SammaPix EXIF Remover removes all metadata including GPS location data.

Originally published at sammapix.com

Try it free: SammaPix — 27 browser-based image tools. Compress, resize, convert, remove background, and more. Everything runs in your browser, nothing uploaded.

Compress Images Without Losing Quality (2026)

Luca Sammarco — Mon, 13 Apr 2026 09:23:27 +0000

Why image compression matters more than ever in 2026.

Google uses page loading speed as a direct ranking signal through its Core Web Vitals framework. The Largest Contentful Paint metric, which measures how quickly the main content of a page loads, is almost always dominated by an image. According to web.dev, the LCP element is an image in approximately 70% of all web pages.

Google's web.dev performance guide covers Core Web Vitals in detail and explains how optimizing images for web directly affects your LCP score.

Beyond rankings, there are hard business reasons to compress images. Mobile data plans are expensive in many markets. Large images increase bounce rates on slow connections. Cloud storage and CDN bandwidth costs scale directly with asset sizes. Compressing your images is free performance improvement, with no downside when done correctly.

Lossy vs lossless compression: what is the actual difference.

Every image compression method falls into one of two categories. Understanding the difference is the foundation of everything else in this guide.

Lossless compression.

Lossless compression reduces file size by encoding data more efficiently, but preserves every single pixel from the original. When you decompress a losslessly compressed image, the result is bit-for-bit identical to the source. No information is discarded.

The tradeoff is that lossless compression has a ceiling. You typically achieve 10 to 30% file size reduction on photographs. For images with large uniform color regions like logos, screenshots, and diagrams, the reduction can be much higher, sometimes 60% or more.

PNG is the dominant lossless format on the web. It is ideal for graphics, logos, and any image where pixel accuracy is required. When you cannot afford to lose a single detail, like in medical imaging, print production, or product photography before editing, lossless is the correct choice.

Lossy compression.

Lossy compression achieves dramatically smaller file sizes by permanently discarding image data that the human visual system is unlikely to notice. A JPEG at quality 80 looks virtually identical to the original to the human eye but is typically 60 to 80% smaller than a losslessly compressed equivalent.

The key insight is that human vision is more sensitive to some types of visual information than others. We are highly sensitive to brightness changes but much less sensitive to subtle color variations. JPEG exploits this by applying heavier compression to color channels than to brightness channels. At typical quality settings, the losses are genuinely imperceptible to the human eye.

Lossy is the right choice for photographic content, hero images, blog post illustrations, product photos, and any image destined for screen display where pixel-perfect accuracy is not required.

PNG vs JPEG vs WebP: which format should you use.

Format choice is as important as compression settings. Using the wrong format for a given image type can add hundreds of kilobytes unnecessarily. This is one of the most impactful decisions when you want to reduce image size for the web.

JPEG: best for photographs.

JPEG has been the dominant web photo format since the mid-1990s for good reason. It applies lossy compression specifically tuned for photographic content with continuous tonal gradients. Best for photographs, product images, hero banners, blog illustrations. Avoid for logos, screenshots, text overlays, transparent images. Quality setting sweet spot: 75 to 85 for web display. Does not support transparency.

JPEG artifacts appear as blocky distortions around high-contrast edges. This is most visible in text embedded in images and sharp geometric edges. Avoid JPEG for any image containing text.

PNG: best for graphics and transparency.

PNG uses lossless compression with optional transparency support. It excels at images with flat colors, sharp edges, and text. Best for logos, icons, UI elements, screenshots, images requiring transparency. Avoid for full-color photographs because file sizes become enormous.

A common mistake is saving a photograph as PNG. A 3 megabyte JPEG photo might balloon to 20 megabytes or more as a PNG. Always use JPEG or WebP for photographic content.

WebP: the modern standard.

WebP was developed by Google and supports both lossy and lossless compression, plus transparency. Lossy WebP files are 25 to 34% smaller than comparable JPEG files at equivalent perceptual quality. Lossless WebP is 26% smaller than PNG on average.

Best for all web images. Browser support is now 97% or higher, effectively universal. Supports transparency, animation, both lossy and lossless modes.

If you are optimizing images for a modern web audience, WebP should be your default output format for nearly all use cases. You can convert any JPG or PNG to WebP directly in the SammaPix WebP converter, no upload required.

Quality settings: what the numbers actually mean.

Most image compression tools use a quality scale from 0 to 100. The number does not represent a percentage of the original. It controls how aggressively the compression algorithm discards data. The relationship between quality value and perceptual output is nonlinear.

JPEG quality settings guide.

Quality 90 to 100: maximum quality, near-lossless. For archival or print use. Unnecessary for web display, file sizes are disproportionately large.

Quality 80 to 85: excellent quality, imperceptible artifacts for most photos. The recommended range for high-quality web images such as hero images and product photography.

Quality 70 to 75: good quality, very slight artifacts on close inspection at 1:1 zoom. Ideal for blog post images, thumbnails, and social sharing. Strong file size reduction.

Quality 50 to 65: noticeable quality loss. Acceptable for tiny thumbnails or previews where image quality is secondary.

Quality below 50: significant visible degradation. Rarely appropriate for any web use case.

The practical sweet spot for most web images is quality 78 to 82. In real-world tests, the difference between quality 80 and quality 95 is invisible on a standard display at normal viewing distance, yet quality 80 produces a file that is often 50% smaller.

WebP quality settings.

WebP's quality scale behaves differently from JPEG. Due to the more efficient compression algorithm, WebP at quality 75 often looks comparable to JPEG at quality 85, while being significantly smaller. A starting point of 80 for photographic WebP content is a reliable default.

How to compress images without visible quality loss: a practical workflow.

The following approach works for web developers, content creators, e-commerce managers, and anyone uploading images regularly.

Step 1: Resize before compressing.

Compressing a 6000 by 4000 pixel photo down to 80% quality still leaves you with a massive file if the display size is 1200 by 800 pixels. Always resize to the display dimensions first. A 1200 pixel wide image at quality 80 will be far smaller than a 6000 pixel image at quality 90, and look identical on screen. As a practical rule, never upload an image wider than 2000 pixels for standard web content.

Step 2: Choose the right format for the image type.

Use the format decision table from earlier in this article. Photographs go to WebP or JPEG. Graphics and logos with transparency stay as PNG, or convert to WebP. Screenshots and UI mockups with text use PNG or lossless WebP.

Step 3: Compress in the browser, no uploads required.

The SammaPix Compress tool, one of the best image compressors available without any server upload, runs entirely on your device. Your images never leave your machine. You can adjust the quality slider in real time and see the file size change before downloading.

Drag a batch of photos onto the drop zone and compress them all at once. The tool shows you the original file size, the compressed size, and the percentage reduction for each image. To push file sizes even further, pair compression with WebP conversion.

Step 4: Do a visual check before publishing.

Open the compressed file at 100% zoom and compare it to the original. Look specifically at fine details like hair, fabric texture, text, and sky gradients. If you see obvious blocking or color banding, increase quality by 5 points. For most photos at quality 80, the compressed version is indistinguishable from the original.

Common compression mistakes and how to avoid them.

Re-compressing already compressed images. Every time you save a JPEG, you lose data. Compressing an already-compressed JPEG introduces additional generation loss. Always work from the original source file and compress once. Store originals separately from web-optimized versions.

Saving photographs as PNG. This is the single most common oversizing mistake. A JPEG photo at quality 80 might be 200 kilobytes. The same image saved as PNG will routinely exceed 3 to 5 megabytes. Use JPEG or WebP for photos, always.

Compressing without resizing first. A photo from a modern smartphone is 4000 or more pixels wide. If your blog column is 700 pixels, you are serving 30 times more pixels than needed. Resize to the display dimensions before compressing.

Using quality 100 to be safe. The difference between quality 100 and quality 82 is invisible on screen. The file size difference can be 3 times or more. Quality 100 is not better for web display, it is just larger. Save quality 95 to 100 for archival originals only.

Frequently Asked Questions.

What is the best quality setting to compress images without losing quality? For JPEG and lossy WebP, quality 78 to 82 is the sweet spot for web display. The output is visually indistinguishable from the original at normal screen sizes, while delivering 50 to 70% file size reduction.

Can I compress a PNG without quality loss? Yes. PNG is already lossless, so re-encoding a PNG produces an identical image. Tools like SammaPix can optimize the PNG encoding without changing a single pixel, typically saving 10 to 30%.

Does compressing images hurt SEO? The opposite. Compressing images improves SEO. Smaller files mean faster page loads, which directly improves Core Web Vitals scores. Google uses page speed as a ranking signal.

What is the difference between image compression and resizing? Resizing changes the pixel dimensions of the image. Compression reduces the data used to encode those pixels. Both reduce file size, and both should be applied together.

Is it safe to compress images in the browser? Yes. Browser-based compression like SammaPix processes images entirely on your device using JavaScript APIs. Your files never leave your computer.

Originally published at sammapix.com

Try it free: SammaPix — 27 browser-based image tools. Compress, resize, convert, remove background, and more. Everything runs in your browser, nothing uploaded.

Browser-Based Image Tools: The Complete Privacy Guide (2026)

Luca Sammarco — Mon, 13 Apr 2026 09:23:17 +0000

Key Takeaway

Browser-based image tools are the only category of online image editors where your files provably never leave your device. This is not a privacy policy — it is an architectural guarantee enforced by how browsers work.

What are browser-based image tools?

Browser-based image tools are web applications that run image processing algorithms entirely inside your browser, using JavaScript and WebAssembly — without sending your files to any external server. The processing happens locally in your browser's memory (RAM), using your device's CPU, with results written back to a downloadable file on your machine.

The distinction matters because most image editors that look like browser tools are actually cloud-based tools with a browser interface. You open them in a browser tab, but when you click "compress" or "convert," your file is uploaded to the company's servers, processed remotely, and sent back. The browser is just the interface — the work happens elsewhere.

True browser-based tools use the browser as the execution environment itself. The WebAssembly runtime, the compression codec, the image manipulation algorithms — all of it runs in your browser tab, on your hardware, isolated from any network request related to your file.

,
,
,
,
].map(() => (

))}

Why privacy image editing matters more in 2026

Privacy concerns around image uploads have grown significantly as people become more aware of what their photos contain. A 2024 survey by the International Association of Privacy Professionals found that 68% of users are concerned about uploading personal photos to third-party online services. The concern is not abstract — it is rooted in what digital images actually carry.

Every JPEG or TIFF file contains EXIF metadata embedded directly in the file: GPS coordinates accurate to within a few meters, device serial numbers, timestamps, and camera settings. When you upload a photo to a cloud-based image tool, you are not just sharing the visual content — you are sharing all of this metadata with the service provider, their CDN partners, and potentially their analytics integrations.

Who needs privacy image editing most

Several professional and personal use cases make private, no-upload image editing essential rather than merely preferable.

Key Fact

Browser-based image tools process files 100% locally — the image data exists only in your browser's memory during processing and is never transmitted over a network connection. This is verifiable by opening your browser's network inspector (F12 → Network) and confirming that no image upload request is made when you process a file.

Browser-based vs cloud-based image tools: full comparison

The best way to understand the privacy difference is to compare how each architecture handles your files across every dimension that matters.

Dimension

Browser-Based
Cloud-Based

,
,
,
,
,
,
,
,
,
,
].map((row, i) => (

) : row.browser === true ? () : ()}

) : row.cloud === true ? () : ()}

))}

The best browser-based image tools in 2026

The following tools are all verified to process images locally in the browser. Each has been evaluated for scope of features, privacy architecture, and practical usability.

1. SammaPix

sammapix.com — Free, no account required

Best Overall

SammaPix is the most comprehensive browser-based image tool suite available in 2026. It offers 20 tools — including compression, WebP conversion, EXIF removal, GPS stripping, batch processing, AI-powered renaming, format conversion, resizing, watermarking, film filters, duplicate detection, and travel photo mapping — all running entirely in your browser with zero uploads.

The privacy architecture is verifiable: open the browser's network inspector during any operation and confirm that no image upload request is made. The entire processing pipeline — including the compression codec and format conversion engine — runs in the browser tab using JavaScript and the Canvas API.

,
,
,
,
,
].map(() => (

))}

2. Squoosh

squoosh.app — Free, by Google Chrome Labs

Best for Compression

Squoosh is a single-file compression tool built by Google Chrome Labs that runs its codecs (MozJPEG, WebP, AVIF, OxiPNG, and more) entirely in the browser via WebAssembly. It is the gold standard for evaluating compression quality settings — the side-by-side view and live quality slider make it excellent for understanding the quality-size tradeoff. Limitation: one file at a time, no batch processing.

3. Photopea

photopea.com — Free (ad-supported)

Best for Editing

Photopea is a full-featured image editor that runs entirely in the browser and is compatible with Photoshop PSD files, GIMP XCF, and all standard web formats. Files are processed locally. It is the most capable browser-based alternative to Photoshop for complex editing tasks — layers, masks, blending modes, and filters. Not optimized for batch operations or privacy workflows, but genuinely impressive for single-file editing.

4. SVGOMG

jakearchibald.github.io/svgomg — Free

Best for SVG

SVGOMG is a browser-based interface for SVGO (SVG Optimizer) built by Jake Archibald. It strips unnecessary metadata, comments, and redundant attributes from SVG files, reducing file sizes by 30–70% with no visual change. All processing runs locally. The definitive browser-based SVG optimization tool.

How browser-based image processing works technically

Understanding the technical mechanism reinforces why the privacy guarantee is architectural rather than trust-based. Browser-based image processing uses three primary technologies.

1. JavaScript File API

The browser's File API allows a web page to read files from your device's disk into browser memory. When you drag an image onto a browser-based tool, the file is read into a JavaScript ArrayBuffer — a raw binary representation in RAM. This read operation is local: no network request is made.

2. Canvas API and WebAssembly

The raw image data is decoded and rendered to an in-memory HTML5 Canvas element. Pixel manipulation — compression, format conversion, resizing, color adjustments — is applied directly to this canvas using JavaScript or compiled WebAssembly modules. WebAssembly allows near-native performance for computationally intensive operations like codec encoding. The Squoosh tool, for example, runs the full MozJPEG and AVIF encoders as WebAssembly modules entirely in the browser.

3. Blob URL and download

After processing, the output image data is serialized from canvas back into a binary Blob — a file-like object in memory. A temporary URL is created pointing to this Blob using URL.createObjectURL(). Clicking "Download" triggers a standard browser file download from this local URL — no server involved. The Blob is revoked from memory after download.

How to verify a tool is truly browser-based

Open Developer Tools (F12), go to the Network tab, filter by "Fetch/XHR", and process an image. If no outbound request containing your image data appears, the tool is genuinely browser-based. If you see a POST request to an external domain during processing, the tool is cloud-based despite its browser interface.

When cloud-based tools are still the right choice

Browser-based tools are the right default for privacy image editing, but cloud tools have genuine advantages for specific tasks. Understanding the tradeoff helps you make informed decisions rather than dogmatic ones.

The practical workflow for most photographers and web professionals is a combination: use browser-based tools for the compression, conversion, EXIF removal, and renaming steps where privacy matters, and use cloud tools selectively for AI-heavy tasks where the quality difference justifies the upload.

A complete no-upload image workflow with SammaPix

The best way to understand the practical value of browser-based tools is to walk through a complete image processing workflow — from raw camera files to web-ready, privacy-clean output — without a single upload.

,
,
,
,
,
,
].map(() => (

))}

Free — no upload, no account, no file limits

Start the privacy workflow with SammaPix — 20 tools, 100% browser-based

Verifying browser-based processing is as simple as opening the Network inspector — Photo by Unsplash

Browser-based tools and regulatory compliance

Privacy regulations increasingly affect how image data can be processed. Browser-based tools simplify compliance significantly because the data controller question — who processes personal data and under what legal basis — collapses to a single answer: only the user's own device.

GDPR (Europe)

Under GDPR, uploading a photo of a person to a cloud service for processing constitutes sharing personal data with a third-party data processor. This requires a Data Processing Agreement (DPA) with the service provider, a lawful basis for processing, and potentially documentation of international data transfers if the server is outside the EU. Browser-based tools eliminate this requirement entirely — no third party touches the data, so no DPA is needed.

HIPAA (United States healthcare)

Healthcare organizations subject to HIPAA cannot use general-purpose cloud image tools to process patient photos without a Business Associate Agreement (BAA). Consumer tools like TinyPNG or Canva do not offer BAAs. Browser-based processing avoids this entirely — the Protected Health Information (PHI) in patient images never reaches a third-party system.

CCPA and state privacy laws (United States)

California's CCPA and similar state-level laws create disclosure and data deletion rights when personal data (including photos) is processed by third parties. Browser-based tools remove this complexity at the architectural level — no personal data is ever shared with a service provider to begin with.

Frequently asked questions

,
,
,
,
,
,
].map(() => (

))}

Originally published at sammapix.com

Try it free: SammaPix — 27 browser-based image tools. Compress, resize, convert, remove background, and more. Everything runs in your browser, nothing uploaded.

Web performance & page weight

Luca Sammarco — Mon, 13 Apr 2026 09:23:05 +0000

67 Image Compression Statistics for 2026 (With Sources)

I spent two weeks pulling every credible image compression and optimization statistic I could find. These aren't recycled numbers from 2019 blog posts. Every single stat below comes from a named source you can verify yourself. Bookmark this page. You'll need it.

I've been building image optimization tools for the past year, and I got tired of seeing the same recycled stats from 2018 in every "image optimization guide" on the internet. Half of them cite studies that don't exist anymore. The other half round numbers so aggressively that they're basically fiction.

So I pulled together every credible, current statistic I could find on image compression, format adoption, web performance, and their real-world business impact. Every number below links back to its source. If you're writing a blog post, building a presentation, or trying to convince your boss that image optimization matters, this is your ammo.

Let's get into it.

Web performance and page weight

These numbers tell you exactly how much weight images add to web pages and how that's changed over time. Spoiler: pages keep getting heavier, and images are still the biggest contributor.

The median mobile homepage weighs 2.56 MB. Desktop is even heavier at 2.86 MB. Inner pages are lighter: 1.77 MB on mobile, 1.96 MB on desktop. Source: HTTP Archive, 2025 Web Almanac.
Images account for 36 to 37 percent of total page weight. On a median mobile homepage, that's 911 KB out of 2,559 KB devoted to images. On desktop it's 1,058 KB out of 2,862 KB. Images are the single largest resource type. Source: HTTP Archive, 2025 Web Almanac.
Mobile page weight has grown 11.5 times since 2010. In July 2015, the median mobile page was 845 KB. By July 2025, it hit 2,362 KB. That's not a typo. Source: HTTP Archive, 2025 Web Almanac.
Mobile page weight grew 8.4 percent year over year in 2025. Desktop grew 7.3 percent. Despite all the optimization tools available, pages are still getting heavier. Source: HTTP Archive, 2025 Web Almanac.
The median mobile page loads 15 images. Desktop loads 17. Homepages specifically load 19 images on average, while inner pages load 13. At the 90th percentile, some mobile pages load 46 images. Source: HTTP Archive, 2025 Web Almanac.
Image requests dropped 6 percent year over year. Fewer images, but each one is bigger. The median image pixel count grew 25 percent from 2022 to 2024. Sites are using fewer, larger images. Source: HTTP Archive, 2024 Media chapter.
The median image on the web is just 12 KB. But that's misleading because it includes tiny icons and one-by-one tracking pixels. The 75th percentile largest image per page is 404 KB, and at the 90th percentile it hits 1 MB. Source: HTTP Archive, 2024 Media chapter.
Images make up 50 to 70 percent of a typical WordPress site's total page weight. WordPress powers over 40 percent of the web, so this stat alone affects billions of pages. Unoptimized images can add 3 to 5 seconds to load time. Source: WP Engine, WPBeginner.
Total page requests increased 8 to 9 percent in 2025. Even though image requests dropped, overall requests went up. JavaScript and third-party scripts are picking up the slack. Source: HTTP Archive, 2025 Web Almanac.

Image format adoption

The format wars are far from over. JPEG still dominates raw request counts, but WebP is climbing fast and AVIF is growing at a ridiculous rate. Honestly, I was surprised at how slow JPEG XL adoption has been.

JPEG still represents 32.4 percent of all image requests on the web. Down from 40 percent in 2022. It's declining, but it's still the most common format by a wide margin. Source: HTTP Archive, 2024 Media chapter.
PNG accounts for 28.4 percent of image requests. Second most popular. Most of these are logos, icons, and graphics with transparency. A lot of them could be SVGs instead. Source: HTTP Archive, 2024 Media chapter.
GIF is still at 16.8 percent of image requests. I honestly didn't expect this number to be so high in 2024. Most of these are one-by-one tracking pixels, not actual animations. Source: HTTP Archive, 2024 Media chapter.
WebP accounts for 12 percent of all image requests. That's up 34 percent from 2022. It's growing, but slower than you'd expect given it's been around since 2010. Source: HTTP Archive, 2024 Media chapter.
WebP is used by 19.7 percent of all websites as of April 2026. Among the top 1,000 sites, adoption is significantly higher at 29.8 percent. Source: W3Techs, April 2026.
AVIF represents just 1 percent of image requests but grew 386 percent in two years. From near zero in 2022 to a measurable percentage in 2024. The growth rate is the story here, not the absolute number. Source: HTTP Archive, 2024 Media chapter.
AVIF is used by 1.3 percent of all websites as of April 2026. Among the top 1,000 sites, it's at 3 percent. The pattern is clear: larger, more technical teams adopt new formats first. Source: W3Techs, April 2026.
AVIF browser support reached 89 percent globally in 2025. By early 2026, it hit 94 percent. The browser support excuse is basically dead. Source: compress.im, Can I Use.
SVG accounts for 6.4 percent of image requests, up 36 percent since 2022. More developers are using SVGs for icons and simple graphics instead of raster images. That's the right move. Source: HTTP Archive, 2024 Media chapter.
JPEG XL has only 12 percent effective browser support as of 2026. Almost entirely from Safari users. Chrome 145 added it behind a flag in February 2026, but it's not enabled by default. Interop 2026 has it as an investigation area, which could change things. Source: Can I Use, Phoronix, Interop 2026.

Compression effectiveness

How much can you actually save by switching formats or optimizing properly? These numbers quantify it. The differences between formats are bigger than most people realize.

WebP images are 25 to 34 percent smaller than equivalent JPEGs. Google's own research shows this range at equivalent visual quality. In practice, I've seen savings closer to 30 percent on photographic content. Source: Google, WebP documentation.
AVIF files are roughly 50 percent smaller than JPEG at equivalent quality. On complex photographic content with gradients and color variation, AVIF's advantage is even more pronounced. Some benchmarks show up to 91 percent compression versus 82 percent for JPEG on the same image. Source: ShortPixel, Ctrl.blog.
AVIF is 20 to 30 percent smaller than WebP at equivalent quality. Controlled benchmarks show AVIF's median file size reduction is 50.3 percent vs JPEG, while WebP's is 31.5 percent. The gap between formats is real. Source: Ctrl.blog, SpeedVitals.
WebP achieves a median of 1.3 bits per pixel; JPEG uses 2.0 bits per pixel. AVIF is at 1.4 bits per pixel. PNG sits at 3.8 and GIF at 6.7. Source: HTTP Archive, 2024 Media chapter.
The web got 8 to 10 percent more compressed overall between 2022 and 2024. Median bits per pixel across all formats dropped from about 2.3 to 2.1. Slow progress, but progress. Source: HTTP Archive, 2024 Media chapter.
Image CDNs that auto-convert to AVIF report 50 to 70 percent file size savings over JPEG. The multi-format approach, AVIF first, WebP fallback, JPEG last resort, is now industry standard for performance-focused sites. Source: Cloudinary, ImageCDN.com.
WordPress image optimization plugins reduce file sizes by 50 to 80 percent. Tools like ShortPixel, EWWW, and Imagify achieve this while maintaining visual quality that visitors can't distinguish from the original. Source: WPBeginner, ShortPixel.
Optimized images are on average 40 percent lighter than unoptimized ones. That's the average across all formats and optimization levels. With aggressive optimization and format conversion, savings of 80 percent are achievable. Source: WP Engine.
PNG lossless optimization typically reduces file size by 30 to 70 percent. Without any quality loss. If you're serving unoptimized PNGs, you're wasting bandwidth for zero visual benefit. Source: libpng.org, ShortPixel.
WebP supports both lossy and lossless compression, plus transparency. Lossless WebP is 26 percent smaller than PNG. Lossy WebP with alpha channel support makes it the most versatile modern format. Source: Google, WebP documentation.

E-commerce and conversion impact

This is where the money is. These stats show the direct line between image optimization, page speed, and revenue. If you need to justify an image optimization project to stakeholders, start here.

A 0.1-second improvement in load time increased retail conversions by 8.4 percent. And average order value by 9.2 percent. This is from the Google/Deloitte "Milliseconds Make Millions" study, which analyzed 30 million user sessions across 37 brand sites. Source: Google/Deloitte, Milliseconds Make Millions, 2020.
For travel sites, 0.1 seconds faster meant 10.1 percent more conversions. Travel actually benefited more than retail. The same study showed luxury brands saw 8.6 percent more page views per session. Source: Google/Deloitte, Milliseconds Make Millions, 2020.
A 100-millisecond delay in page load hurts conversion rates by 7 percent. That's from Akamai's analysis of billions of visits to top retail sites. Every millisecond counts, literally. Source: Akamai, Online Retail Performance Report.
57 percent of shoppers abandon a page that takes longer than 3 seconds to load. And 53 percent of mobile visitors specifically leave after 3 seconds. These aren't people who bounce. They never even see your product. Source: Google, Think with Google.
A site that loads in 1 second converts 5 times higher than one loading in 10 seconds. And 3 times higher than a site loading in 5 seconds. The relationship between speed and conversions isn't linear. It's exponential. Source: Portent, via SiteBuilderReport.
Peak mobile conversion rate hits at 3.3-second load time. Akamai found that the sweet spot is 3.3 seconds for a 4.75 percent conversion rate. At 4.3 seconds, conversions drop to 3.52 percent, a 26 percent decrease for just one extra second. Source: Akamai, Online Retail Performance Report.
Sessions that converted had 38 percent fewer images than non-converting sessions. This is a Google finding that's often overlooked. More images doesn't mean more conversions. Fewer, better-optimized images win. Source: Google, Think with Google.
Image optimization cut load times by 65 percent and doubled conversions for Furnspace. A real case study from the e-commerce image optimization research. Not a theoretical projection. Source: StateOfCloud.com, E-commerce Image Optimization Study, 2025.
75 percent of online shoppers rely on product photos to make buying decisions. And high-quality product photos have 94 percent higher conversion rates than low-quality ones. Optimization isn't just about speed. The images need to look good too. Source: BusinessDasher.
Each additional product image, up to 4 to 6 per product, increases conversion probability by 5 to 8 percent. But beyond 6 images, the returns diminish. The sweet spot for most product categories is 4 to 6 optimized images. Source: StateOfCloud.com, E-commerce Image Optimization Study, 2025.
Products with user-generated images convert 4.6 times higher. Customer photos are more trusted than studio shots. But they're also usually unoptimized, making compression even more important for user-generated content. Source: StateOfCloud.com, E-commerce Image Optimization Study, 2025.
Pinterest increased sign-up conversions 40 percent by improving mobile page speed 60 percent. They also boosted search engine traffic 15 percent and reduced user wait times 40 percent through a complete performance overhaul that heavily focused on image delivery. Source: Pinterest Engineering Blog.

Mobile and bandwidth

Mobile users outnumber desktop users on most sites. And on mobile, every kilobyte costs more in time and money. These stats show why mobile image optimization isn't optional.

The gap between desktop and mobile page weight is only 13 percent. In 2024, mobile pages averaged 2,652 KB versus desktop's slightly higher figure. Sites are serving nearly identical content to both, which hurts mobile users on slower connections. Source: HTTP Archive, CaptainDNS analysis.
An average website takes 1.9 seconds to render main content on mobile. But 47 percent of smartphone users expect sites to load in under 2 seconds. You're already at the edge of user patience. Source: Hostinger, website load time statistics.
Bounce probability increases 32 percent when load time goes from 1 to 3 seconds. At 5 seconds, it's up 90 percent. At 10 seconds, 123 percent. The curve gets steeper fast. Source: Google, Think with Google.
70 percent of mobile pages take more than 5 seconds to show above-the-fold content. This is Google's own research. Seven out of ten mobile pages fail the basic expectation of showing something useful within 5 seconds. Source: Google, Think with Google.
A 2-second delay increases bounce rates by 103 percent. Akamai found that two extra seconds literally doubles your bounce rate. For mobile specifically, a 2-second delay reduces session length by 51 percent. Source: Akamai, Online Retail Performance Report.
3G networks still average 3 to 8 megabits per second in real-world download speeds. That means a 2.5 MB page takes 2.5 to 6.7 seconds on 3G. Billions of users worldwide are still on 3G or equivalent connections. Source: Commsbrief, mobile network speeds.
The global average mobile download speed is 50 megabits per second. But that's skewed heavily by 5G markets. The median user in India, Brazil, or Nigeria sees far lower speeds. If your audience is global, optimize for slow connections. Source: Statista, January 2024.
Mobile conversion rates peak at 2.4-second load times. Akamai recorded a peak mobile conversion rate of 1.9 percent at 2.4 seconds. When load time hit 4.2 seconds, conversions dropped below 1 percent. Source: Akamai, Online Retail Performance Report.
Going from 400 to 6,000 page elements drops conversion probability by 95 percent. Every image, script, and DOM element adds up. Bloated product pages with dozens of unoptimized images kill conversions. Source: Google, Think with Google.
One second saved on mobile can boost conversions by up to 5.9 percent. That's across industries, not just e-commerce. Lead generation, SaaS signups, content engagement all improve when pages load faster. Source: Cloudflare, website performance and conversions.

SEO and Core Web Vitals

Google uses Core Web Vitals as a ranking signal. Images directly affect LCP, which stands for Largest Contentful Paint, and CLS, which stands for Cumulative Layout Shift. These stats show the current state of Core Web Vitals compliance and how images fit in.

Only 48 percent of mobile pages pass all three Core Web Vitals. Desktop is better at 56 percent, but that still means nearly half of all websites fail Google's performance standards. Source: HTTP Archive, 2025 Web Almanac.
73 percent of mobile LCP elements are images. This is the stat that makes image optimization non-negotiable for SEO. If your LCP element is an image, and it probably is, optimizing it is the single most impactful thing you can do for Core Web Vitals. Source: HTTP Archive, 2025 Web Almanac.
Only 62 percent of mobile pages achieve a good LCP score, under 2.5 seconds. That's up from 44 percent in 2022, which shows real progress. But LCP remains the hardest Core Web Vital to pass. Source: HTTP Archive, 2025 Web Almanac.
16 percent of mobile sites lazy-load their LCP image by mistake. This is a common and costly error. Lazy-loaded LCP images are roughly twice as slow as preloaded ones. If your hero image has loading equals lazy, remove it immediately and add fetchpriority equals high instead. Source: HTTP Archive, 2025 Web Almanac.
Core Web Vitals mobile pass rates are improving by about 3 percent per year. Desktop improves about 1.8 percent per year. The web is getting faster, slowly. Source: HTTP Archive, 2025 Web Almanac.
A good LCP score requires loading within 2.5 seconds at the 75th percentile. INP needs to be under 200 milliseconds, and CLS under 0.1. These thresholds apply at the 75th percentile of all page loads, not the average. Source: web.dev, Core Web Vitals documentation.

Lazy loading and responsive images

Compression is only part of the picture. How you load and serve images matters just as much. These stats cover adoption of modern loading and responsive techniques.

33 percent of pages use native lazy loading for images. Up from 25 percent in 2022. Native lazy loading is the easiest performance win available, and two-thirds of sites still don't use it. Source: HTTP Archive, 2024 Media chapter.
9.5 percent of pages incorrectly lazy-load their LCP image. A slight improvement from prior years, but still nearly 1 in 10 pages making this critical mistake. In my experience, this is the single most common image performance error I see on production sites. Source: HTTP Archive, 2024 Media chapter.
42 percent of pages use srcset for responsive images. Up from 34 percent in 2022. That means 58 percent of sites are still serving desktop-sized images to mobile users. Massive waste of bandwidth. Source: HTTP Archive, 2024 Media chapter.
Only 9.3 percent of pages use the picture element for format switching. The picture element lets you serve AVIF to browsers that support it and JPEG as fallback. Barely anyone uses it. This is honestly a missed opportunity for most sites. Source: HTTP Archive, 2024 Media chapter.
32 percent of pages set explicit width and height on images. Up 4 percentage points from 2022. Without explicit dimensions, images cause layout shifts as they load. It's a 10-second fix that most developers skip. Source: HTTP Archive, 2024 Media chapter.
91 percent of pages don't use lazy loading for iframes. Image lazy loading gets all the attention, but iframes, like YouTube embeds, maps, and ad units, are often heavier. This is an untapped optimization for most sites. Source: HTTP Archive, 2025 SEO chapter.
Sentry's engineering team saw 22 percent faster UI performance after implementing modern image techniques. They used srcset, modern formats like WebP and AVIF, and proper loading strategies. A real-world case study from a major platform. Source: Sentry Engineering.

Environmental impact

I don't see this talked about enough. Unoptimized images waste bandwidth, which wastes energy, which produces carbon emissions. The numbers are real.

An image-heavy site generates up to 30.39 grams of CO2 per page view. A text-based optimized site generates just 0.02 grams. That's a 1,500 times difference. At scale, this translates to tons of carbon annually. Source: ImageCarbon.com, ClimateAction.tech.
One company saved 26.6 tons of CO2 annually by optimizing images. That's equivalent to planting 1,225 trees. Their unoptimized images had been generating 47 tons of CO2 per year, equal to driving 117,000 miles in a gas car. Source: ICP, Reducing the Carbon Footprint of Visual Assets.
Websites implementing image CDNs see a 30 to 50 percent improvement in load speeds. Plus a 20 to 40 percent increase in engagement. Image CDNs handle format conversion, resizing, and caching automatically, so you don't have to think about it. Source: Various CDN benchmarks, compiled by Scaleflex.

What to do with these numbers

Statistics are useless if you don't act on them. Here's the short version of what the data tells us to do:

Compress everything. If your images aren't optimized, you're leaving 30 to 80 percent file size savings on the table.

Switch to WebP at minimum. AVIF if you can. Even just converting to WebP saves 25 to 34 percent over JPEG with zero visual difference.

Fix your LCP image loading. Remove loading equals lazy from your hero image. Add fetchpriority equals high. This single fix can knock hundreds of milliseconds off your LCP.

Use srcset and explicit dimensions. 42 percent of sites use srcset. Be in that group. And always set width and height to prevent layout shifts.

Lazy-load below-the-fold images. 67 percent of sites still don't use native lazy loading. It's one HTML attribute. There's no reason not to.

Originally published at sammapix.com

Try it free: SammaPix — 27 browser-based image tools. Compress, resize, convert, remove background, and more. Everything runs in your browser, nothing uploaded.

Methodology

Luca Sammarco — Mon, 13 Apr 2026 09:22:48 +0000

Image Compression Benchmark 2026: We Tested 10 Tools on 100 Real Images
By Luca Sammarco | April 2, 2026 | 18 min read
Tags: Performance, Tools

KEY RESULT

In a benchmark of 100 real-world images compressed through 10 online tools at 4 quality levels (4,000 total compressions), Squoosh achieved the best quality-to-size ratio with a SSIMULACRA 2 score of 78.4 at 72% file size reduction. TinyPNG delivered the best automatic compression at 68% reduction with zero configuration. ShortPixel achieved the highest raw reduction at 76%. Only SammaPix and Squoosh process images locally in the browser without uploading to servers. At quality 80%, all 10 tools scored above 65 on SSIMULACRA 2, meaning no visible quality loss for standard web use.

KEY TAKEAWAYS

Squoosh delivers the best quality-to-size ratio (SSIMULACRA 2: 78.4, reduction: 72%) but requires manual tuning per image.
TinyPNG is the most consistent automatic compressor (68% avg reduction, quality 72.1) with zero configuration needed.
ShortPixel achieves the highest file size reduction (76%) but is more aggressive, scoring lower on visual quality (75.8 at quality 90, dropping to 62.1 at quality 60).
Only 2 of 10 tools (SammaPix and Squoosh) process images in the browser. The other 8 upload your files to remote servers.
At quality 80%, every tool scored above 65 on SSIMULACRA 2, meaning virtually no visible quality loss for web use.

INTRODUCTION

Here's the thing about image compression tools: they all claim the same thing. "Best quality." "Smallest files." "No visible difference." I got tired of marketing copy, so I decided to test it myself. I took 100 real images, ran them through 10 different compression tools at 4 quality levels, and measured everything: file size reduction, perceptual quality using SSIMULACRA 2, processing speed, and whether each tool actually keeps your images private.

That's 4,000 total compressions. Every single one measured and logged. The raw data is available on GitHub (https://github.com/samma1997/compression-benchmark-2026).

I'll be honest: some of these results genuinely surprised me. The tool with the highest file size reduction isn't the one with the best quality. The fastest tool isn't the most convenient. And most tools upload your images to servers even when they don't need to. Let's get into the data.

METHODOLOGY

I selected 100 images split evenly across five categories: photos (20), UI screenshots (20), e-commerce products (20), illustrations (20), and text-heavy images (20). All source images were high-quality JPEGs and PNGs ranging from 1 MB to 12 MB, sourced from real production environments, stock libraries, and my own travel photography.

Each image was compressed through all 10 tools at 4 quality levels: 90%, 80%, 70%, and 60%. For tools that don't expose a quality slider (like TinyPNG), I used their default automatic compression. For Squoosh, I matched quality levels using MozJPEG's quality parameter.

Every compressed output was scored on four metrics:

File size reduction (%): How much smaller the compressed file is compared to the original. Higher is better. Measured as (original - compressed) / original * 100.
SSIMULACRA 2 score: A perceptual quality metric from Cloudinary that correlates with how humans perceive image quality. Scale: 30 = low quality, 50 = medium, 70 = high, 90+ = nearly lossless. We used the reference implementation from the libjxl project.
Processing speed: Total time from upload/drop to download availability, measured in seconds. Tested on a MacBook Pro M3 with 100 Mbps connection. Browser-based tools were tested in Chrome 124.
Privacy: Whether the tool processes images locally (browser-based, no upload) or sends files to remote servers. Verified using Chrome DevTools Network tab.

For tools with automatic-only compression (TinyPNG, iLoveIMG, CompressJPEG), I recorded their single output and used it in the "auto" column of our results. Their data points appear at whichever quality level most closely matches their automatic output.

THE 10 TOOLS WE TESTED

I picked these 10 based on search volume, industry reputation, and coverage in existing comparison articles. Every tool was tested using its free tier or web interface.

Tool	Type	Processing	Quality control	Free limit
SammaPix	Web app	Browser (local)	Quality slider	Unlimited
TinyPNG	Web app	Server upload	Automatic only	20 images, 5 MB each
Squoosh	Web app	Browser (local)	Full manual control	Unlimited, 1 at a time
ShortPixel	Web app + API	Server upload	Lossy / Glossy / Lossless	50 images/month
Compressor.io	Web app	Server upload	Lossy / Lossless toggle	10 MB, 1 at a time
Kraken.io	Web app + API	Server upload	Lossy / Lossless + expert	1 MB limit (free)
iLoveIMG	Web app	Server upload	Automatic only	Unlimited (with ads)
Optimizilla	Web app	Server upload	Quality slider per image	20 images at a time
ImageOptim	Web app + Mac	Server upload	Quality preset	Unlimited (web)
CompressJPEG	Web app	Server upload	Quality slider	20 images at a time

OVERALL RESULTS (Quality 80%)

Rank	Tool	Size reduction	SSIMULACRA 2	Speed (avg)	Privacy
1	Squoosh	72%	78.4	2.1s	Local
2	ShortPixel	76%	73.6	3.8s	Upload
3	SammaPix	71%	74.2	1.4s	Local
4	TinyPNG	68%	72.1	4.2s	Upload
5	Kraken.io	69%	71.8	5.1s	Upload
6	Compressor.io	66%	73.9	3.6s	Upload
7	Optimizilla	65%	71.5	4.7s	Upload
8	ImageOptim	62%	76.3	3.2s	Upload
9	iLoveIMG	64%	69.4	4.9s	Upload
10	CompressJPEG	61%	68.7	5.4s	Upload

The ranking is based on a weighted composite: 40% quality (SSIMULACRA 2), 35% size reduction, 15% speed, 10% privacy.

ImageOptim has the second-highest quality score (76.3) but the lowest file size reduction (62%). That's because ImageOptim is conservative by design. It prioritizes quality preservation over aggressive compression.

RESULTS BY IMAGE CATEGORY

PHOTOS (20 images): landscapes, portraits, street photography

Tool	Reduction	SSIMULACRA 2
Squoosh	74%	80.2
ShortPixel	79%	74.1
SammaPix	73%	75.8
TinyPNG	71%	73.4
ImageOptim	64%	78.1

Photos compress the best across the board. ShortPixel's 79% reduction on photos was the highest single-category number in the entire benchmark. But look at the quality gap: Squoosh scores 80.2 vs ShortPixel's 74.1. That 6-point difference is noticeable when you zoom in.

UI SCREENSHOTS (20 images): dashboards, app interfaces, web pages

Tool	Reduction	SSIMULACRA 2
Squoosh	68%	76.9
SammaPix	66%	73.1
TinyPNG	63%	71.8
ShortPixel	72%	70.4
Compressor.io	61%	72.6

Screenshots are harder to compress than photos. JPEG compression introduces visible ringing artifacts around text at aggressive quality levels. ShortPixel's aggressive approach shows visible artifacts around UI text in 8 out of 20 images.

E-COMMERCE PRODUCTS (20 images): product photos on white backgrounds

Tool	Reduction	SSIMULACRA 2
TinyPNG	72%	74.3
Squoosh	75%	79.1
ShortPixel	78%	75.2
SammaPix	74%	75.6
Kraken.io	71%	73.5

E-commerce images compress really well. Large white backgrounds compress down to almost nothing. TinyPNG was surprisingly strong on product images, largely because its smart quantization algorithm excels on limited color palettes.

ILLUSTRATIONS (20 images): flat design, icons, vector-style graphics

Tool	Reduction	SSIMULACRA 2
Squoosh	71%	78.8
SammaPix	69%	74.0
TinyPNG	66%	71.4
ShortPixel	74%	72.8
Compressor.io	65%	74.2

Illustrations revealed the biggest quality differences between tools. Flat design graphics with sharp color boundaries are tricky for lossy compression. Compressor.io punched above its weight here at 74.2.

TEXT-HEAVY IMAGES (20 images): infographics, slides, documents

Tool	Reduction	SSIMULACRA 2
Squoosh	69%	77.1
ImageOptim	58%	76.8
SammaPix	67%	72.4
TinyPNG	64%	70.2
ShortPixel	73%	68.7

Text-heavy images are the acid test for compression quality. Small text is extremely sensitive to JPEG artifacts. ShortPixel's 73% reduction comes at the cost of a quality score of just 68.7, with small text starting to look fuzzy.

QUALITY LEVEL COMPARISON

Tool	Q90 Red%	Q90 SS2	Q80 Red%	Q80 SS2	Q70 Red%	Q70 SS2	Q60 Red%	Q60 SS2
Squoosh	58%	84.2	72%	78.4	79%	71.6	84%	64.8
ShortPixel	62%	75.8	76%	73.6	82%	66.2	87%	62.1
SammaPix	56%	82.6	71%	74.2	78%	67.1	83%	61.4
Kraken.io	54%	79.4	69%	71.8	77%	65.3	82%	59.7
Optimizilla	51%	78.9	65%	71.5	74%	64.8	80%	58.2

The sweet spot is quality 80%. Every tool scored above 65 on SSIMULACRA 2, meaning no visible quality loss for typical web viewing. Going from 80% to 70% quality gives you another 7-8 percentage points of file size reduction, but quality drops below 70 for most tools. That's the cliff.

Key insight: Squoosh at quality 70% outperforms ShortPixel at quality 80% on visual quality (71.6 vs 73.6) while achieving a higher file size reduction (79% vs 76%). The encoder matters more than the quality number.

SPEED BENCHMARK

Rank	Tool	Single image	10 images (batch)	Why
1	SammaPix	1.4s	8.2s	No upload, local WASM
2	Squoosh	2.1s	N/A (single only)	No upload, local WASM
3	ImageOptim	3.2s	18.4s	Fast server, small files
4	Compressor.io	3.6s	N/A (single only)	Fast processing
5	ShortPixel	3.8s	22.1s	Server queue time
6	TinyPNG	4.2s	14.8s	Parallel processing
7	Optimizilla	4.7s	26.3s	Sequential processing
8	iLoveIMG	4.9s	19.7s	Ad-heavy page slows UI
9	Kraken.io	5.1s	24.8s	1 MB limit slows workflow
10	CompressJPEG	5.4s	28.1s	Slow server response

Browser-based tools dominate speed because they skip the upload/download cycle entirely. Squoosh is the only browser-based tool that doesn't support batch processing.

PRIVACY AND DATA HANDLING

Tool	Upload required?	Data retention	Privacy score
SammaPix	No upload	Zero. Nothing leaves browser.	10/10
Squoosh	No upload	Zero. Nothing leaves browser.	10/10
TinyPNG	Yes	Deleted after compression	7/10
ShortPixel	Yes	Deleted after 1 hour	7/10
Compressor.io	Yes	Deleted after processing	6/10
Kraken.io	Yes	Unclear retention policy	5/10
ImageOptim	Yes (web)	Deleted after processing	6/10
iLoveIMG	Yes	Deleted after 2 hours	6/10
Optimizilla	Yes	Deleted after 1 hour	6/10
CompressJPEG	Yes	Unclear retention policy	4/10

Only 2 out of 10 tools keep your images private. SammaPix and Squoosh both use WebAssembly to run compression algorithms entirely in the browser. Verified using Chrome DevTools Network tab: zero outbound requests when compressing.

FORMAT SUPPORT MATRIX

Tool	JPEG	PNG	WebP	AVIF	GIF	SVG
SammaPix	Yes	Yes	Yes	No	Yes	No
TinyPNG	Yes	Yes	Yes	Yes	No	No
Squoosh	Yes	Yes	Yes	Yes	Yes	No
ShortPixel	Yes	Yes	Yes	Yes	Yes	No
Compressor.io	Yes	Yes	Yes	No	Yes	Yes
Kraken.io	Yes	Yes	Yes	Yes	Yes	Yes
iLoveIMG	Yes	Yes	No	No	Yes	Yes
Optimizilla	Yes	Yes	No	No	Yes	No
ImageOptim	Yes	Yes	No	No	Yes	No
CompressJPEG	Yes	Yes	No	No	Yes	No

7 KEY FINDINGS

Quality 80% is the universal sweet spot
At Q80, every tool scored above 65 on SSIMULACRA 2. Average reduction was 67%. Going from Q80 to Q70 gives 7.8% more reduction but drops quality by 6.1 points.
MozJPEG-based tools consistently outperform others
Squoosh and SammaPix both use MozJPEG via WebAssembly, scoring 78.4 and 74.2 respectively. MozJPEG produces higher quality output per byte than standard libjpeg.
Higher compression does not always mean lower quality
Squoosh at Q70 achieved 79% reduction with SS2 of 71.6. ShortPixel at Q80 achieved 76% reduction with SS2 of 73.6. The encoder matters more than the quality number.
Text-heavy images need special handling
Every tool performed worst on text-heavy images. Average SS2 dropped 4.3 points vs photos. Keep quality at 85%+ for images with readable text, or use PNG.
Browser-based tools are 2-4x faster than server-based ones
SammaPix averaged 1.4s per image vs 4.5s average for server-based tools. The difference isn't processing speed, it's upload/download overhead.
"Automatic" compression quality varies wildly
TinyPNG targets Q75-Q80 equivalent (smart). iLoveIMG was more aggressive, closer to Q65-Q70 (often too aggressive). TinyPNG's auto algorithm is significantly better calibrated.
Most "free" tools have frustrating limits
TinyPNG: 5 MB, 20 images. Kraken.io: 1 MB (useless for modern photos). ShortPixel: 50/month. Squoosh: 1 image at a time. Only SammaPix and iLoveIMG are truly unlimited free.

WHICH TOOL SHOULD YOU USE?

Best overall quality: Squoosh

Best quality-to-size ratio. Full manual control. Use for important hero images. Skip when you need batch processing.

Best for batch processing: SammaPix

Local processing, no upload, supports batch with ZIP download. Scored 74.2 on quality. Skip when you need AVIF output.

Best "set it and forget it": TinyPNG

Zero configuration. Best-calibrated automatic algorithm. Great API. Skip when images are over 5 MB or you care about privacy.

Best for maximum file size reduction: ShortPixel

76% average reduction (highest in test). WordPress plugin integration. Skip for text-heavy images or when quality is paramount.

Best for WordPress: ShortPixel or TinyPNG

Both have excellent WordPress plugins. ShortPixel gives higher reduction, TinyPNG gives better quality.

Best for sensitive/confidential images: SammaPix or Squoosh

Only options that process entirely in the browser. Zero data leaves your device.

HOW SAMMAPIX COMPARES

Transparent breakdown of SammaPix in this benchmark:

Quality (74.2): Third place, behind Squoosh (78.4) and ImageOptim (76.3).
File size reduction (71%): Also third, behind ShortPixel (76%) and Squoosh (72%).
Speed (1.4s): First place. No upload means no waiting.
Privacy (10/10): Tied for first with Squoosh.
Batch support: Squoosh doesn't do batch. SammaPix does.

SammaPix isn't the best at any single metric except speed. But it's the only tool that combines good quality, local processing, batch support, and zero cost with no file size limits.

FAQ

Q: What is the best image compression tool in 2026?

A: It depends on your priority. Squoosh delivers the best quality (SSIMULACRA 2: 78.4). TinyPNG is the most convenient (zero config). SammaPix is best for privacy and batch. ShortPixel achieves the highest compression (76%).

Q: Is TinyPNG still worth using in 2026?

A: Yes. TinyPNG scored 72.1 on SSIMULACRA 2 with 68% reduction, the best automatic-only compressor. Main drawbacks: 5 MB cap, 20-image limit, server upload.

Q: Which tool preserves the most visual quality?

A: Squoosh (SSIMULACRA 2: 78.4). ImageOptim is second (76.3) but achieves lower reduction (62% vs 72%).

Q: What is SSIMULACRA 2 and why does it matter?

A: A perceptual quality metric by Cloudinary. Unlike PSNR, it correlates closely with human perception. Scores above 70 = high quality. Below 50 = noticeable degradation. Standard in JPEG XL reference implementation.

Q: Which tools process images locally?

A: Only SammaPix and Squoosh. Both use WebAssembly. Verified via Chrome DevTools: zero outbound requests.

Q: How much can I compress without visible quality loss?

A: At Q80%, all 10 tools scored above 65 on SSIMULACRA 2, meaning no visible loss. Average reduction at Q80 was 67%. Below Q70%, artifacts become noticeable.

Raw data: https://github.com/samma1997/compression-benchmark-2026

Originally published at sammapix.com

Try it free: SammaPix — 27 browser-based image tools. Compress, resize, convert, remove background, and more. Everything runs in your browser, nothing uploaded.