Test release checked
What does this image compressor do?
Vaultaire compresses one static JPEG, PNG, or WebP inside your browser. Choose an encoder quality or a maximum file-size ceiling, compare the result, and download only after the new file decodes and its size, dimensions, transparency, and supported metadata are checked.
The original file is not changed. JPEG and WebP use the browser's lossy encoder. PNG uses a fresh lossless re-encode of the browser-rendered pixels, which may not make the file smaller.
How do I compress an image under a file-size limit?
- Choose one static image. The browser checks its signature, dimensions, animation, orientation, transparency, and supported metadata.
- Enter the ceiling. Choose JPEG or WebP, select Maximum file size, and enter KB or MB. The exact interpreted byte count appears before compression.
- Review the verified copy. Vaultaire reports whether the ceiling was met. It never changes dimensions to force the result.
A 2 MB ceiling means 2,000,000 bytes. It is not an exact-size request. If no tested quality fits, the page shows the smallest verified candidate and links to the image resizer.
Compress an image to 20 KB, 50 KB, 100 KB, or 200 KB
Upload forms often state a hard limit: 20 KB for an exam portal, 50 KB or 100 KB for a job application, 200 KB for a government form, 1 MB or 2 MB for a general upload. Select JPEG or WebP output, select Maximum file size, and type the number with its unit.
Vaultaire reads the ceiling in decimal bytes, so 50 KB means 50,000 bytes and 200 KB means 200,000 bytes. It then tests a bounded set of encoder qualities and keeps the highest one whose real output is at or below your number. The reported size is the size of the actual file, measured after encoding, not an estimate.
The ceiling is a limit, not a target. The tool will not pad a file to hit an exact number, and it never reduces dimensions to force a fit. A result under your ceiling is a pass.
Small targets fail more often than large ones, and the reason is pixel count. A 12-megapixel phone photo carries far more detail than 20 KB of JPEG can hold at any quality the encoder offers. When the tool reports that no tested quality fits, the fix is to shrink the dimensions first: open the image resizer, reduce the width, then return here and set the ceiling again. Tight targets such as 20 KB and 50 KB usually need that step. Targets of 200 KB and above usually do not.
What image compression changes
Lossy JPEG and WebP encoding can change color values, fine texture, sharp edges, and text. A numeric quality value controls one browser encoder; it does not mean the same visual quality in every browser or release.
The displayed width and height stay fixed. Stored EXIF orientation is applied to the pixels once, then removed. The new file is rendered through an 8-bit sRGB canvas, so wide-gamut, HDR, CMYK, higher-bit-depth, density, and print color behavior may change.
The browser APIs used for local output are documented by OffscreenCanvas.convertToBlob and HTMLCanvasElement.toBlob.
JPEG, PNG, or WebP output
| Format | Compression | Transparency | Maximum-size mode |
|---|---|---|---|
| JPEG | Lossy | No. A background choice is required. | Yes |
| PNG | Lossless from rendered pixels | Yes | No |
| WebP | Lossy | Yes, after an alpha check | Yes |
The browser tests each encoder after file selection. A format is disabled if its test output, signature, or second decode does not match.
What happens to EXIF, GPS, copyright, and C2PA data?
Compression creates a new image and does not copy supported source metadata. The inspection view lists supported EXIF, GPS, XMP, IPTC, copyright, authorship, editing, origin, and C2PA findings before you continue. The output is scanned again.
Removing authorship and provenance may be the wrong choice for a work you publish professionally. Keep the original. For metadata removal without changing encoded pixels, use the photo metadata remover and EXIF viewer.
The supported scan cannot rule out every proprietary metadata field, invisible pixel watermark, steganographic signal, or later platform label.
Private image compression without an upload
Cloudflare serves this page and its static code before you select a file. After selection, the bytes move only between the page and a same-origin worker in your browser. Client analytics send fixed action events to Vaultaire's PostHog proxy, the only connection the route's Content Security Policy allows, and the image bytes are never part of them.
A later navigation still makes a normal web request. JavaScript also cannot promise secure erasure of browser memory, so the page does not make that claim.
Image compressor FAQ
Are my images uploaded?
No. The selected image is read, compressed, and verified inside your browser, with no upload or processing API. PostHog receives fixed action and completion events, never image bytes, filenames, or settings. The route's Content Security Policy allows only Vaultaire's PostHog proxy for those events.
How do I compress an image under 2 MB?
Choose JPEG or WebP output, select Maximum file size, enter 2 MB, and compress. Vaultaire tests a bounded set of encoder settings and reports whether the exact output bytes meet the ceiling.
Can the tool make an image exactly 100 KB?
No exact-size promise is possible. The setting is a ceiling. The tool looks for the highest tested quality at or below 100,000 bytes and says plainly when compression alone cannot reach it.
How do I compress an image to 50 KB?
Choose JPEG or WebP output, select Maximum file size, and enter 50 KB. Vaultaire reads that as 50,000 bytes, tests a bounded set of encoder qualities, and keeps the highest one whose real output is at or below the ceiling. Targets this small often fail on a full-resolution phone photo, because a 12-megapixel image holds more detail than 50 KB can carry. When no tested quality fits, reduce the width in the image resizer first, then set the ceiling again. The same steps apply to a 20 KB, 100 KB, or 200 KB target.
How do I reduce image size without losing quality?
Choose PNG output. PNG is lossless here, so the pixels are preserved exactly, though a lossless re-encode may not make the file smaller. Truly lossless compression and a large size reduction are not available at the same time. For a smaller file with no visible difference, choose WebP at a high quality setting and compare the before and after view at full size before you download. WebP usually reaches a smaller file than JPEG at a matching quality.
Does image compression reduce dimensions?
No. This tool keeps the displayed width and height unchanged. Use the separate image resizer when fewer pixels are required.
Will JPEG or WebP compression reduce quality?
Lossy JPEG and WebP encoding can change pixel values and visible detail. The quality value controls the browser encoder; it is not a universal visual-quality score.
Does compression remove EXIF and GPS metadata?
The new copy does not copy supported EXIF, GPS, XMP, IPTC, origin, or C2PA records. Vaultaire scans the output again, but it cannot rule out every proprietary or pixel-embedded signal.
Can I compress several images at once?
Not in this test release. It processes one image at a time so memory use, cleanup, and every receipt remain clear.
What happens to PNG transparency?
PNG and WebP output must preserve the source alpha values during local verification. JPEG has no transparency, so the tool requires a background choice before it can create a JPEG.
What happens when the compressed image would be larger?
Vaultaire discards the larger candidate and does not offer it for download. Keep the original, choose another format or setting, or use the image resizer.
Can it compress HEIC, AVIF, GIF, or animated images?
No. This release accepts one static JPEG, PNG, or WebP. It rejects animation instead of exporting only the first frame.
Does it preserve print color and DPI?
No print or archival guarantee is made. Browser canvas encoding can convert color, bit depth, profiles, and density metadata. Keep the original and use a color-managed desktop workflow for print masters.