UtilityKit

500+ fast, free tools. Most run in your browser only; Image & PDF tools upload files to the backend when you run them.

Image Compressor

Compress images with quality control

About Image Compressor

Phone cameras produce 8 MB photos that take six seconds to load on 4G. Oversized images are the fastest fix for slow pages and bounced email attachments. This tool sends your JPEG, PNG, or WebP through Sharp — powered by mozjpeg, pngquant, and libwebp — and returns the smallest file that still looks identical at normal viewing distance. A 4.2 MB iPhone photo routinely lands under 420 KB at quality 75 with no perceptible loss. Output JPEGs are encoded as progressive so browsers paint a preview first, then sharpen — improving Largest Contentful Paint scores. Files are deleted from the server within seconds. No storage, no analytics, no watermarks, and no per-day quota — compress a hundred product photos in a row without hitting a paywall.

Why use Image Compressor

Up to 90% File Size Reduction

A 4.2 MB JPEG from a smartphone routinely lands under 420 KB at quality 75 with no perceptible loss. Sharp's mozjpeg encoder squeezes out bytes that hand-tuned encoders miss, achieving reductions that desktop apps and CMS auto-optimisers rarely match.

Quality Slider 1–100

Default 80 is the web sweet spot — drop to 70 for thumbnails or 90 for full-bleed hero images. Visual difference is barely perceptible above quality 75, while file size differences between 75 and 95 are often greater than 50%.

JPEG, PNG, and WebP Supported

JPEG uses mozjpeg, PNG uses pngquant lossy quantisation, and WebP uses libwebp — each format gets the best-in-class compressor rather than one generic encoder. This means meaningfully better results than apps that treat every format identically.

Progressive JPEG Output

Output JPEGs are encoded as progressive, painting a blurry preview instantly and sharpening as more data arrives. This improves Largest Contentful Paint scores on slow connections because something visible appears before the full file is downloaded.

Auto-Deleted After Processing

Files are processed inside a temporary directory on the backend and deleted within seconds of returning the result. There is no storage bucket, no analytics on your images, and no third-party CDN receiving your photos.

No Quota, No Watermarks

Compress 100 product photos back-to-back without hitting a daily cap or a watermark prompt. No account is required, and the tool never adds a visible or invisible overlay to your output images.

How to use Image Compressor

  1. Drop a JPEG, PNG, or WebP file (up to 25 MB) onto the upload area or click to browse
  2. Pick your output format — keep the original, convert to JPEG, PNG, or WebP
  3. Drag the quality slider between 1 and 100; the default of 80 is the recommended web sweet spot
  4. Click Compress and watch the progress indicator
  5. Compare the before and after file sizes and inspect the side-by-side preview
  6. Click Download to save the compressed file

When to use Image Compressor

  • Before uploading product photos to Shopify, WooCommerce, or any ecommerce platform that serves images byte-for-byte
  • When a blog post scores a poor Lighthouse score because hero images exceed 500 KB
  • Before emailing a batch of event or real-estate photos that bounce off a 25 MB attachment limit
  • Before pushing design assets to a GitHub repo where large binaries inflate clone times
  • When a web form rejects an image upload because the file exceeds its size limit
  • Before attaching screenshots to a bug report or support ticket in a project management tool

Examples

Phone photo (JPEG)

Input: iPhone 14 photo: 4032×3024, 4.2 MB JPEG, camera quality 96

Output: Compressed JPEG: 4032×3024, 420 KB, quality 75 — 90% smaller, visually identical at normal viewing distance

Product shot (PNG → WebP)

Input: Studio PNG with transparency: 2000×2000, 3.8 MB

Output: WebP output: 2000×2000, 240 KB, quality 80 — 94% smaller, transparency preserved

Hero banner (PNG → WebP)

Input: Marketing PNG: 1920×1080, 2.1 MB

Output: WebP output: 1920×1080, 95 KB, quality 80 — 95% smaller, ideal for above-the-fold load

Tips

  • Quality 75 is the web sweet spot — most viewers cannot tell the difference compared to quality 95, and the file is typically 60% smaller.
  • For photos, JPEG and WebP both use lossy compression. For screenshots and logos with text, use PNG or lossless WebP to keep edges crisp.
  • If you plan to crop or edit further, compress only after the final edit — re-encoding a lossy format multiple times compounds quality loss each time.
  • WebP saves roughly 30% over JPEG at the same visual quality but is not fully supported in older email clients — safe for web, risky for email newsletters.
  • Run images through this tool before uploading to a CMS. Most content management systems serve the original byte-for-byte without optimising on their end.

Frequently Asked Questions

How much can I shrink a JPEG before it looks bad?
Quality 75 is the web sweet spot — most viewers cannot tell the difference compared to quality 95, and the file is typically 50–60% smaller. Below quality 60 you start to see blocking artefacts and colour banding, especially around high-contrast edges like text on a background.
Does compression remove EXIF or GPS data from my photos?
Re-encoding through Sharp strips most EXIF metadata as a side effect, including GPS coordinates. If you need to keep colour profile data (ICC) intact for print work, choose PNG output, which preserves the embedded profile.
What's the difference between lossy and lossless compression?
Lossy compression (JPEG and WebP with quality < 100) permanently discards imperceptible colour data to achieve large size reductions. Lossless compression (PNG, lossless WebP) reorganises data mathematically without discarding anything — the decoded pixels are bit-identical to the source, but size reductions are more modest.
Will a 10 MB phone photo compress to under 1 MB?
Yes, routinely. A 4032×3024 JPEG from an iPhone at default quality 80 typically comes out between 350 KB and 600 KB — a reduction of 85–95%. The exact result depends on image content: flat sky and plain surfaces compress more than detailed foliage or textured fabric.
Can I compress a PNG with transparency without losing the alpha channel?
Yes. Choosing PNG output preserves the alpha channel through pngquant's lossy palette quantisation. If you convert a transparent PNG to JPEG, the alpha is dropped and replaced with a white background — choose the output format deliberately.
Why is WebP usually smaller than JPEG at the same quality?
WebP uses a more modern prediction algorithm (derived from VP8/VP9 video coding) that models how adjacent pixels relate. JPEG's DCT-based encoding was designed in 1992. At the same visual quality setting, WebP typically saves 25–35% over JPEG.
Are my photos stored on your server?
No. Files are written to a temporary directory for the duration of processing and deleted immediately after the compressed file is returned to your browser. Nothing is stored in a database, CDN, or object storage bucket.
Is there a maximum file size I can upload?
The upload limit is 25 MB per file. Files above this threshold are rejected before processing. Most phone photos are under 12 MB, so this limit is rarely hit in practice.

Explore the category

Glossary

Quality factor
A 1–100 integer that controls how aggressively lossy encoders discard visual data. Higher values preserve more detail but produce larger files; values below 75 begin to show visible artefacts in most photos.
mozjpeg
An open-source JPEG encoder maintained by Mozilla that applies additional optimisation passes beyond the baseline JPEG standard, achieving 10–15% smaller files at equivalent visual quality compared to libjpeg.
Lossy vs lossless
Lossy compression permanently discards imperceptible data to achieve large file reductions; decoded pixels differ slightly from the source. Lossless compression reorganises data without any information loss, producing identical pixels on decode.
Progressive JPEG
A JPEG encoding mode that stores the image in multiple low-to-high frequency passes. Browsers can render a blurry preview before the full file loads, improving perceived performance on slow connections.
Chroma subsampling
A JPEG and WebP encoding technique that stores colour (chroma) at lower resolution than brightness (luma), exploiting the fact that human vision is more sensitive to brightness differences than colour differences.
Largest Contentful Paint
A Core Web Vitals metric measuring how long it takes the largest visible element — often a hero image — to finish rendering. Reducing image file size directly improves this score.