Reduce Image File Size Online
Learn how to shrink image files online while preserving quality, choosing the right format, dimensions, and compression settings.
To reduce an image file online, resize it to the dimensions you actually need, choose a suitable format, then increase compression gradually while checking the result at its intended display size. Keep the original file, because every lossy export can discard image data.
Quick workflow
- Save an untouched copy of the original.
- Resize the pixel dimensions to match the upload, website, message, or document where the image will be used.
- Choose JPEG or WebP for most photographs, PNG or lossless WebP for graphics that need sharp edges or transparency, and AVIF when the destination supports it.
- Adjust quality in small steps. Compare file size and visible detail after every export.
- Inspect text, line art, gradients, transparency, and high-contrast edges at the intended viewing size.
- Check the service’s upload, retention, and privacy terms before sending sensitive images.
What actually makes an image file smaller?
An image file usually becomes smaller through one or more of these changes:
| Change | What it does | Best use | Risk |
|---|---|---|---|
| Reduce dimensions | Stores fewer pixels | A photo displayed in a small card or message | Upscaling later can look soft |
| Lossy compression | Removes visual information | Photographs and other detailed images | Blocking, ringing, blur, or banding |
| Lossless compression | Re-encodes without changing image data | Logos, diagrams, screenshots, archival copies | Usually produces a larger file than lossy compression |
| Format conversion | Uses a codec better suited to the image | WebP or AVIF where supported | Older destinations may not accept newer formats |
| Metadata removal | Removes EXIF and other non-pixel data | Web publishing or privacy-sensitive sharing | GPS, camera, copyright, or orientation data may be lost |
Lossy compression can be effective for photos but makes artifacts more visible around sharp edges, line art, and text. Lossless compression preserves the original image data. web.dev’s image-performance guidance recommends experimenting with settings and judging the actual output rather than relying on one universal quality value.
Resize before compressing
Dimensions are width and height in pixels. A 4,000 × 3,000 image contains far more pixel data than a 1,200 × 900 image, even if both depict the same scene. Export at the largest size the destination actually needs, with a little allowance for high-density displays.
- Profile photo: use the service’s documented maximum instead of uploading the camera original.
- Blog or product image: match the largest rendered width, then account for retina displays if needed.
- Email or messaging: choose a practical viewing size rather than preserving print dimensions.
- Print: do not reduce dimensions solely to save bytes unless the printer’s required resolution permits it.
Do not repeatedly resize the same derivative. Return to the original and create a new export when you need another size.
Choose JPEG, PNG, WebP, or AVIF
| Format | Use it for | Advantages | Watch for |
|---|---|---|---|
| JPEG | Photographs and continuous-tone images | Broad compatibility and effective lossy compression | No transparency; artifacts increase at low quality |
| PNG | Text, diagrams, screenshots, and transparency | Lossless encoding and alpha transparency | Often much larger than a well-compressed photo format |
| WebP | Web photos, graphics, transparency, and animation | Supports lossy and lossless compression, transparency, and animation | Confirm that the destination accepts it |
| AVIF | Modern web delivery where supported | Often compresses better than older formats | Compatibility and tooling vary by destination |
Google documents WebP’s lossy, lossless, transparency, and animation support in its WebP format guide. MDN’s image-format guide explains the compatibility tradeoffs among JPEG, PNG, WebP, and AVIF. RFC 9649 describes WebP as supporting both lossy and lossless compression, alpha transparency, and animation: RFC 9649.
How to use an online image compressor
- Open a compressor that states which formats and maximum file sizes it accepts.
- Upload a copy, not your original.
- Set output dimensions if the tool provides resizing.
- Select the output format. Keep PNG for sharp graphics or required transparency; use JPEG, WebP, or AVIF for photographs when supported.
- Start with moderate compression. Export, download, and compare.
- Increase compression only until artifacts become visible, then move one step back.
- Open the downloaded file and confirm dimensions, orientation, transparency, animation, and color appearance.
Online services differ in upload limits, retention, privacy, pricing, and supported formats. The available guidance names tools such as Squoosh, ImageOptim, and image-optimization services, but does not establish current policies for any individual provider. Review the selected service’s documentation before uploading confidential files.
Compress locally when the image is sensitive
Local processing avoids uploading the source file to a third party. A desktop editor, command-line encoder, or browser-based local tool can resize and export the image without sending it to a server. Keep the original, export a derivative, and compare the result at the intended display size.
# Example with ImageMagick (run locally)
magick input.jpg -resize 1600x1600\> -strip -quality 82 output.jpgThe exact command and available codecs depend on your installed ImageMagick build. The -strip option removes metadata, so omit it when metadata is required.
How much quality should you sacrifice?
There is no quality number that works for every image. A detailed photo may tolerate lossy compression well, while a screenshot containing small text may become unreadable quickly. Test several nearby settings and compare:
- Fine text and one-pixel lines
- High-contrast boundaries and diagonals
- Faces, hair, foliage, and other detailed textures
- Gradients and shadows, where banding can appear
- Transparent edges and semi-transparent overlays
Judge the image at the size people will see it. A 200% inspection can reveal problems, but it should not force you to keep unnecessary bytes when the artifact is invisible at the intended size.
Edge cases
Transparency
JPEG cannot store transparency. Use PNG, WebP, or AVIF when the transparent background must remain. Check whether the destination renders alpha correctly.
Animated images
Do not convert an animated image to a still format unless losing animation is acceptable. WebP supports animation; verify that the receiving application supports animated WebP.
Screenshots and text-heavy images
PNG or lossless WebP usually preserves crisp text better than aggressive JPEG compression. If the screenshot is photographic and contains little text, a moderate lossy setting may be reasonable.
EXIF orientation and metadata
Some images rely on an orientation tag instead of physically rotating pixels. After processing, verify that the image is upright. Removing metadata can also remove GPS, camera, copyright, or color-profile information.
Color profiles
Images can look different when a tool removes or mishandles color profiles. Compare the output in the same viewer and keep a color-managed workflow for design or print work.
Very large uploads
If an online tool rejects the file, resize locally first or use a service with a documented higher limit. Do not repeatedly re-save a partially compressed derivative; start again from the original.
Troubleshooting
| Symptom | Likely cause | Fix |
|---|---|---|
| File is still too large | Dimensions are excessive or compression is lossless | Reduce dimensions, switch to a suitable lossy format for photographs, or remove unnecessary metadata. |
| Text looks blurry | Lossy compression damaged sharp edges | Use PNG or lossless WebP, increase quality, and export from the original. |
| Blocky squares or ringing appear | Compression is too aggressive | Raise quality gradually and inspect high-contrast edges. |
| Transparent background becomes white | Output format does not support alpha or the export flattened it | Use PNG, WebP, or AVIF and verify alpha settings. |
| Image is rotated | Orientation metadata was ignored or removed | Apply rotation to pixels before stripping metadata, then verify the export. |
| Colors changed | Color profile handling differs between tools | Preserve the profile when required and compare in a color-managed viewer. |
| Online upload is refused | File-size, dimension, format, or policy limit | Check the service documentation, resize locally, or choose a supported format. |
| Download is smaller but unusable | The wrong format or an overly low quality setting was selected | Return to the original, choose a format suited to the content, and test a less aggressive setting. |
Performance, reliability, and cost considerations
- Performance: resizing first reduces upload time and processing work. For websites, deliver an appropriately sized responsive image instead of one oversized source.
- Reliability: retain originals and record the output dimensions, format, and quality settings so you can reproduce a derivative.
- Privacy: local processing keeps sensitive files on your device. For online tools, read current retention and deletion terms before uploading.
- Cost: many basic compressors are free, but limits, queues, batch processing, and premium formats vary. Compare the total workflow cost, including storage and bandwidth.
- Compatibility: modern formats can reduce bytes, but keep a fallback when the destination has older format requirements. web.dev’s format guidance covers this decision.
Or skip the browser setup
If your workflow first needs a clean screenshot of a web page before you optimize or store it, ScreenshotNeo provides a single-request capture API. Cookie and consent banners, newsletter popups, and chat widgets are removed before the shot. Bot checks, blank pages, failed loads, timeouts, and cache hits are not billed, and the response identifies the result with X-Page-Verdict and X-Billed headers. Its MCP server lets Claude, Cursor, and other MCP clients call take_screenshot, get_page_info, and capture_pdf.
See the ScreenshotNeo API documentation for all options.
curl -G "https://api.screenshotneo.com/v1/shot" -d access_key=YOUR_API_KEY --data-urlencode url=https://stripe.com -o shot.webpimport requests
r = requests.get("https://api.screenshotneo.com/v1/shot", params={"access_key": "YOUR_API_KEY", "url": "https://stripe.com"}, timeout=90)
open("shot.webp", "wb").write(r.content)const q = new URLSearchParams({ access_key: 'YOUR_API_KEY', url: 'https://stripe.com' });
const res = await fetch(`https://api.screenshotneo.com/v1/shot?${q}`);The Free plan includes 1,000 screenshots each month with no card. Paid plans start at $5 for 3,000 shots, and every feature is included on every plan. Create a free ScreenshotNeo account.
FAQ
Can I reduce an image without visible quality loss?
Yes, when resizing removes pixels that are not needed or when lossless compression is sufficient. Lossy compression always trades some image information for fewer bytes, so inspect the export.
Is WebP always smaller than JPEG?
No format wins for every image and setting. Compare outputs at the dimensions and quality you need, and confirm destination support.
Should I compress before or after resizing?
Resize first, then compress the final dimensions. Export again from the original when creating a different size.
What is the safest way to shrink a private photo?
Process it locally or use an online service whose current upload, retention, and deletion terms meet your requirements.
Why did a PNG become larger after compression?
PNG is lossless, so it may not shrink a photographic image much. Use a suitable photographic format when transparency and exact pixel preservation are not required.
