Image Compressor

Why compress images?

You have a 4 MB photo and your CMS upload limit is 1 MB. Your email attachment bounces because the image is too large. Your page is slow because the hero image weighs more than every script on the page combined. These are the everyday problems that image compression solves.

The practical reasons to reduce image size before sharing or publishing are well established:

• Faster page loads and better Core Web Vitals. Large images are the single most common cause of slow Largest Contentful Paint. Compressing a photo from 4 MB to 400 KB cuts a significant fraction of your page weight.
• Email attachment limits.Most mail providers cap attachments at 10–25 MB. A compressed image rarely triggers these limits.
• Mobile data. Visitors on cellular connections pay for every byte. A 300 KB photo loads in under a second on 4G; a 4 MB photo does not.
• Upload size caps. CMSes, social platforms, and form endpoints frequently enforce maximum file sizes. Compressing before upload eliminates silent failures and rejections.
• Storage cost. If you store thousands of images on a cloud bucket, halving their average file size halves your storage bill.

Lossy vs lossless — why this tool is lossy

Image compression falls into two categories, and the distinction matters for choosing the right tool.

Lossless compression (used by PNG and GIF) preserves every pixel exactly. The file gets smaller through clever encoding — run-length encoding, dictionary compression — but the decoded image is bit-identical to the original. Lossless compression has a ceiling. A screenshot of a flat UI compresses well; a photograph of a forest barely compresses at all, because the random-looking pixel variation in natural images defeats the patterns that lossless encoders exploit.

Lossy compression(JPG, WebP, AVIF) discards perceptually-irrelevant detail. Files get dramatically smaller — often 10× smaller — at the cost of subtle quality loss. JPG and WebP achieve this by discarding what your eyes cannot see: high-frequency color noise, slight gradients in low-contrast areas, microtexture in skin and sky. The result looks nearly identical to the original at typical viewing sizes and distances.

This tool is lossy. It re-encodes your image as JPG or WebP at the quality level you choose. PNG is intentionally not an output option because PNG cannot compress photographic content meaningfully — for a photo input, converting to JPG or WebP first is the right move and is exactly what this compressor does.

What the quality slider actually does

The quality number (1–100) is not a "percentage of original quality preserved." It is an encoder parameter that controls how aggressively the lossy compression discards detail. Understanding it helps you pick the right value without guessing.

• 100 — the encoder discards almost nothing. Files are the largest; quality is nearly identical to the source. Rarely worth using for a compression workflow.
• 80 (the default)— the encoder discards a lot of detail you cannot see at normal viewing sizes. Files are typically 5–10× smaller than at quality 100. This is the sweet spot for almost every use case. The image compression online industry default is 80 for a reason.
• 60 — subtle artifacts start to appear in flat areas like sky and skin. Still acceptable for thumbnails, previews, and non-critical images.
• 40— visible blocking artifacts in detailed areas. Use only when file size matters more than quality — for example, a small thumbnail viewed at 80×80 pixels where artifacts are invisible at that size.
• 10 — heavy degradation. Useful only for extremely small placeholder previews.

This tool re-encodes the image as you drag the slider, updating the output file size and visual preview in real time. The practical workflow: drag the slider down until you see artifacts you find unacceptable, then back up one notch. That is your optimal setting for this image.

JPG vs WebP — which output format?

Both JPG and WebP are lossy formats, but they behave differently in practice:

• JPG — universal compatibility. Every device, every browser, every email client, every CMS handles JPG. Slightly larger files than WebP at the same quality. The right choice when you are not sure where the image will be used.
• WebP— roughly 25–35% smaller than JPG at the same quality. Universally supported in modern browsers since 2020. Use for any image you control on a website. Older email clients and some legacy CMSs still do not accept WebP — check your destination before using it for non-web contexts.
• Auto mode — this tool picks WebP if your source was WebP, JPG otherwise. This is the safe default. The source format usually tells you what the destination expects.

To compress PNG, the tool converts it to JPG or WebP on output. There is no PNG output option because PNG re-compression produces no meaningful size reduction for photographic content. If you specifically need PNG-to-PNG lossless optimization (for UI screenshots or logos), use a dedicated tool such as pngquant or oxipng — those are not browser tools and are out of scope here.

What this tool does — and does not — do

A precise accounting of scope, so there are no surprises:

• Re-encodes images as JPG or WebP at a chosen quality level, reducing file size via lossy image compression.
• Live file-size preview as you adjust the quality slider.
• Preserves the source image dimensions exactly — no resizing. Use the image resizer for that.
• Strips EXIF metadata on re-encode. Privacy win for sharing online; downside for photo workflows that depend on camera or GPS data.
• Does not do PNG-to-PNG lossless re-compression.
• Does not do AVIF output — browser encode support is still limited; will revisit in a future update.
• Does not batch-process multiple files. One image at a time.

How to use this image compressor

1. Drag an image onto the drop zone, or click to browse. PNG, JPG, and WebP are accepted as input.
2. Pick the output format. Auto is the safe default — it matches the source format where possible, or selects JPG for PNG inputs.
3. Drag the quality slider. The output file size updates in real time as you move it. The visual preview updates alongside it.
4. Start at 80 (the default). Drag lower if the file is still too large; back up when you see artifacts you cannot accept.
5. When the file size and visual quality look right, click Download to save the compressed image to your device.

Example

A typical phone photo compressed for a web page:

Input:   4032 × 3024 px  •  JPG  •  4.1 MB
Format:  Auto (JPG)
Quality: 80

Output:  4032 × 3024 px  •  JPG  •  380 KB  (−91%)

The 91% reduction is typical for a phone photo at quality 80. The dimensions are unchanged — only the encoding quality is reduced. At typical screen sizes the output is visually indistinguishable from the original.

Input:   1920 × 1080 px  •  PNG  •  2.8 MB (screenshot)
Format:  WebP
Quality: 80

Output:  1920 × 1080 px  •  WebP  •  210 KB  (−93%)

Note: converting a screenshot with text from PNG to WebP at quality 80 may introduce subtle ringing artifacts around sharp edges and letters. For screenshots of UI or text, keeping the file as PNG and reducing resolution or cropping is generally a better approach. Use this compressor for photographic content, not for screenshots.

Privacy and security

Your image never leaves your browser. Compression happens entirely on your device using the browser's Canvas API — the pixel data is drawn onto a canvas element and re-encoded to a Blob at the quality you select. There is no upload, no server-side processing, and no copy stored anywhere.

Most "compress image online" tools send your file to a server, run the encoding on a backend, return the result, and retain a temporary copy during processing. That copy may be logged or left on disk longer than the privacy policy implies. If you are compressing internal screenshots, client photos, proprietary product images, or anything you would not paste into a public web form, the browser-local approach is categorically safer.

No analytics event records the file content or filename. No signup is required.

Frequently asked questions

Why can I not output PNG from this compressor?

PNG is a lossless format — it cannot meaningfully compress photographic content the way JPG and WebP can. A photo re-saved as PNG is roughly the same size as the original PNG, because the encoder cannot discard any pixel data. Compressing a PNG photo to a smaller file requires converting to a lossy format. This tool does that automatically by only offering JPG and WebP as outputs. If you specifically need smaller PNG files for UI screenshots or logos with transparency, use a dedicated lossless PNG optimizer such as pngquant.

My PNG screenshot of text looks blurry after compression.

JPG and WebP are tuned for photographs and natural images, not for text and sharp geometric edges. They produce visible artifacts around high-contrast transitions — the "ringing" effect around letters is a classic symptom. For screenshots of UI or code, keep them as PNG. The file size can be reduced by cropping, lowering resolution, or using a PNG-specific lossless optimizer. Converting a text screenshot to JPG or WebP is almost always the wrong move.

What is the right quality to start with?

80%. Always start at 80 and inspect the result. This is the industry default for JPG and WebP compression for a reason: it is the point where most users cannot distinguish the output from the original at typical viewing sizes. If the file is still too large at 80, try 70. If you see unacceptable artifacts at 80, try 85. Very few images require going below 70 or above 85.

What does "Saved 82%" actually mean?

The output file is 82% smaller than the input. So if you started with a 1.4 MB photo and saved 82%, the result is roughly 252 KB. The percentage shown is the reduction in byte count, not a quality rating. A high savings percentage does not necessarily mean the image looks bad — at quality 80, 80–90% size reduction is normal and the visual result is typically excellent for photographs.

Does compression strip EXIF metadata?

Yes. The Canvas API discards all EXIF data — camera model, exposure settings, GPS coordinates, capture timestamps — when re-encoding the image. For sharing images online this is a privacy benefit: you are not accidentally publishing location data embedded in a phone photo. If you need EXIF to travel with the file (photography workflows, archiving, print), use a desktop tool such as ImageMagick or a dedicated EXIF editor to reattach the metadata after compression.

Can I recompress an already-compressed JPG?

Yes, but each lossy re-encode adds another generation of artifacts. Starting from a high-quality JPG source and re-encoding at quality 80 typically produces a good result. Starting from an already heavily compressed JPG and re-encoding loses more detail. For best results, always compress from the highest-quality source available — ideally the original file from the camera or design tool — rather than re-compressing an already-compressed version.

Is my image data sent anywhere?

No. The compression runs entirely in your browser using the Canvas API. No file is uploaded, no server processes your image, and no copy is stored anywhere outside your device. You can safely reduce image size for files containing sensitive internal content — they will never be transmitted over the network.