Image Compressor
Compress PNG, JPG and WebP images with a quality slider and live before/after canvas comparison. Target file size mode, batch compress with ZIP download, per-image quality control, WebP output, EXIF strip. Images are processed in your browser — never uploaded to any server.
Image Compressor Tool
Everything in this free image compressor
Built to outperform TinyPNG, Squoosh and Optimizilla on the features that matter — without paywalls, file limits or server uploads.
How to compress images online — step by step
How this image compressor compares to alternatives
Most free image compressors upload your files to a server, limit batch size, or lock key features behind paid accounts. LazyTools processes everything in your browser with no restrictions.
| Feature | LazyTools ✦ | TinyPNG | Squoosh | Optimizilla |
|---|---|---|---|---|
| Quality slider | ✔ Yes | ✘ No | ✔ Yes | ✔ Yes |
| Live before/after canvas comparison | ✔ Draggable slider | ✔ Yes | ✔ Yes | ✔ Yes |
| Target file size mode | ✔ Yes | ✘ No | ✘ No | ✘ No |
| Batch compression | ✔ 50 images | ✔ 20 (free) | ✘ Single only | ✔ 20 images |
| ZIP download for batch | ✔ Free | ✘ Paid only | ✘ No | ✘ No |
| Per-image quality in batch | ✔ Yes | ✘ No | ✘ No | ✔ Yes |
| WebP output from any format | ✔ Yes | ✔ Paid only | ✔ Yes | ✘ No |
| EXIF metadata strip | ✔ Yes | ✔ Auto | ✘ No | ✘ No |
| Max dimension constraint | ✔ Yes | ✘ No | ✘ No | ✘ No |
| Images never uploaded to server | ✔ Always | ✘ Uploads to server | ✔ Yes | ✘ Uploads to server |
| No account required | ✔ Yes | ✔ Yes | ✔ Yes | ✔ Yes |
| No watermark on output | ✔ Never | ✔ Yes | ✔ Yes | ✔ Yes |
| Monthly image limit (free) | ✔ Unlimited | 100/month | ✔ Unlimited | ✔ Unlimited |
The Complete Guide to Image Compression — JPG, PNG, WebP and File Size Reduction
Image compression is one of the most impactful optimisations for website performance. Images are the single largest contributor to page weight on most websites, and uncompressed images are the most commonly flagged issue in Google PageSpeed Insights. A 2 MB hero image compressed to 200 KB loads ten times faster, directly improving Core Web Vitals scores, search ranking and user experience — particularly on mobile connections.
How image compression works
Digital images are made of pixels, each storing colour information. Compression algorithms reduce file size by encoding this information more efficiently. Two approaches exist: lossy and lossless.
Lossy compression permanently discards pixel data that the human eye is unlikely to notice — typically by averaging similar colour values in adjacent pixels or by removing high-frequency detail in smooth areas. JPEG uses lossy compression, and the quality slider controls how aggressively information is discarded. At 85% quality, most images lose around 60% of their file size with no perceptible quality degradation. At 60% quality, file sizes drop by 80–85% but compression artefacts become visible in smooth gradients.
Lossless compression reorganises pixel data to represent it more efficiently without discarding anything. PNG uses lossless compression, which means every pixel in a PNG is stored exactly — but file sizes are significantly larger than an equivalent JPEG. Lossless compression cannot meaningfully reduce PNG file size beyond what the algorithm already achieves; the only way to reduce a PNG is to convert it to a lossy format.
JPEG compression — the sweet spot
JPEG is the most widely used format for photographs and complex images with gradients and continuous tone. The compression algorithm divides the image into 8×8 pixel blocks, applies a discrete cosine transform to each block, and quantises the result. At high quality settings, fine detail is preserved. At lower settings, blocks become visually distinct — the characteristic "blocky" JPEG artefact appearance.
For web images, the widely accepted optimum is 75–85% quality. At 80%, a typical 3 MB photograph compresses to 300–500 KB — a 83–90% file size reduction — with no visible quality difference at normal viewing size. At 70%, the same photograph is 150–250 KB with mild artefacts in smooth background areas. The quality slider in this tool lets you find exactly the right balance for each image by comparing the before and after side by side at full resolution.
WebP — why you should convert your images in 2025
WebP is an image format developed by Google in 2010 and now supported by all major browsers including Safari (since 2020). It uses more advanced compression algorithms than JPEG, typically achieving 25–35% smaller file sizes at equivalent perceptual quality. For a 500 KB JPEG, an equivalent quality WebP is typically 330–375 KB.
WebP also supports transparency — unlike JPEG — making it a direct replacement for PNG in most web use cases. A PNG graphic with transparency that weighs 200 KB can often be compressed to a 50–80 KB WebP with lossless quality and full transparency preserved. The only remaining advantage of PNG is guaranteed pixel-perfect fidelity, which matters for screenshots containing text and programmatically generated diagrams.
To convert your images to WebP using this tool, simply select WebP as the output format. The compression engine handles the conversion automatically, regardless of whether your input file is JPG, PNG or GIF.
PNG compression — what actually works
PNG uses lossless DEFLATE compression — the same algorithm used in ZIP files. The PNG compressor finds repeated pixel patterns and encodes them more compactly. Because no data is discarded, there is a ceiling to how much this reduces file size. A PNG that has already been optimised by any decent tool has very little headroom for further lossless compression.
Many users try to compress PNG files and are surprised when tools seem to have little effect. The reason is that PNG lossless compression is already close to optimal for most images. If you need a significantly smaller file from a PNG source, the correct approach is to convert to WebP (which uses lossy compression while preserving transparency) or JPG (if the image has no transparency). The output format selector in this tool handles this conversion automatically.
EXIF metadata and file size
EXIF data is metadata embedded in image files by cameras, smartphones and photo editing software. A typical smartphone photo contains 40–100 KB of EXIF metadata including GPS coordinates, altitude, camera manufacturer and model, focal length, shutter speed, ISO, lens details, software version, and the date and time the photo was taken.
Stripping EXIF reduces file size by 20–80 KB depending on what was embedded. More importantly, EXIF removal is a privacy essential for anyone sharing photos online. GPS coordinates in EXIF data reveal the exact location where a photo was taken — including the home address if the photo was taken indoors. Sharing images with EXIF intact on social media, forums or public websites exposes this data to anyone who downloads the file.
The Strip EXIF toggle in this tool is enabled by default. Disabling it preserves the metadata — useful for photographers maintaining archival metadata workflows or sending work to clients who need shooting data.
Target file size compression — a smarter approach
Many platforms have specific upload limits: email services typically restrict attachments to 5–10 MB; some social platforms scale images to a maximum resolution; e-commerce platforms specify maximum product image file sizes. Manual quality adjustment to hit a target often requires multiple attempts.
The target file size mode in this tool automates this process. Enter your maximum file size in KB, and the compressor uses a binary search algorithm to find the highest quality setting that produces an output within your target. The result is the best possible quality at your specified size constraint — in a single operation, without trial and error.
Image compression and Core Web Vitals
Google's Core Web Vitals metrics directly measure page loading experience and are a confirmed ranking factor in Google Search. The Largest Contentful Paint (LCP) metric measures how long it takes for the largest visible element — usually a hero image or featured photo — to fully render. Slow LCP scores are most commonly caused by uncompressed, oversized images.
A hero image compressed from 2 MB to 200 KB loads ten times faster on the same connection. On a typical 4G mobile connection (25 Mbps), that is the difference between 640ms and 64ms transfer time — before any other network overhead. Google recommends serving images in next-gen formats (WebP, AVIF), compressing images appropriately, and specifying image dimensions to prevent layout shift. This tool addresses the first two recommendations directly.
When not to compress images
Compression always involves a trade-off. Avoid aggressive compression for images that will be further edited or printed — always compress from the original source file, never re-compress an already-compressed image, as each compression pass multiplies artefacts. Medical imaging, forensic photography, fine art reproductions and archival photography require lossless formats or very high quality settings (95%+) where visual fidelity is non-negotiable. For these use cases, PNG or high-quality TIFF remains the correct choice.