Save JPG photos as lossless PNG files for editing, layering and design work. Convert one image or a whole batch, privately in your browser.
No limits or paywall.
Use instantly, no account.
Processed inside your browser.
Clean, full-quality output.
No upload queue or waiting.
Works on phone and desktop.
JPG is the default for cameras and phones because it makes photos small, but that efficiency comes from throwing information away every time the file is saved. Open a JPG, add a caption, save, crop, save again, and the image slowly degrades: edges get softer, colours band, and tell-tale 8 × 8 blocks appear. Converting to PNG ends that cycle. PNG is lossless, so the image you have now is exactly the image you will have after a hundred more edits.
This converter turns JPG and JPEG files into PNG in your browser. It is free, needs no account, adds no watermark and works for single images or large batches. Use it whenever you are about to edit, layer, annotate or publish an image where crisp detail matters more than file size.
Before
JPG, quality 85, 1920 × 1080
≈ 400 KB
Compact, but every re-save in an editor adds a little more compression damage.
After
PNG, 24-bit lossless
≈ 2.5–3.5 MB
Larger on disk, but the pixels are now frozen: edit and re-save without further loss.
Converting to PNG cannot restore detail the JPG already discarded. It stops further loss. Typical sizes for a photo; a graphic or screenshot saved as JPG converts to a much smaller PNG.
Drop one or more JPG or JPEG files into the workspace at the top of the page, or click to choose them from your device.
PNG is preselected for this page. The quality slider does not apply because PNG is always lossless.
Click Convert Now. Each image is decoded and re-encoded as PNG locally on your device.
Save individual PNGs or use Download All Converted to get a single ZIP archive.
| Property | JPG | PNG | Practical takeaway |
|---|---|---|---|
| Compression | Lossy (DCT) | Lossless (DEFLATE) | PNG never degrades on re-save |
| Photo file size | 200–600 KB | 2–4 MB | Expect PNG photos to be larger |
| Screenshot / text size | Often larger, with halos | Small and razor-sharp | PNG wins for UI and text |
| Transparency | Not supported | 8-bit alpha channel | Add transparency later in an editor |
| Editing round-trips | Quality drops each save | No generation loss | Use PNG as your working copy |
| Colour depth | 8 bits per channel | 8 or 16 bits per channel | 16-bit helps heavy colour grading |
| Compatibility | Universal | Universal | Both work everywhere |
When you convert, the JPG is fully decoded into raw RGB pixels, the same pixels you see on screen. Those pixels are then written into a PNG container using DEFLATE compression, which can reproduce them exactly. Nothing is added and nothing more is removed. The visual result is identical to the JPG on screen; the difference is entirely in what happens next.
Because PNG is lossless it has to preserve the noise in the photo and the faint compression artefacts the JPG introduced. Those patterns are effectively random, and random data does not compress well, so a photographic PNG is usually several times larger than its JPG source. That is normal. For screenshots and graphics that were mistakenly saved as JPG, the opposite can happen: once converted, the flat areas compress efficiently and the PNG may be a similar size, while the edges stay sharp from here on.
Each time a JPG is saved, the encoder re-runs its lossy steps: colour subsampling, block transforms and quantisation. Even at the same quality setting the result drifts slightly from the previous save because the input already contains artefacts. After a few rounds of edit and save, the damage becomes visible as smeared detail and blotchy skies. Photographers call this generation loss.
The professional workflow is simple: keep a lossless master (PNG or TIFF), make all your edits on that master, and export a JPG only at the very end for distribution. Converting incoming JPGs to PNG before you start editing puts you on that workflow even if the original came from a phone or a client email.
JPG approximates each 8 × 8 block with a sum of smooth wave patterns. Smooth photographs are described well with just a few of those waves, but a hard edge, such as black text on a white background, needs many high-frequency components. When quality settings round those components away, you see ghost outlines around letters. PNG has no such limitation: an edge is just two pixel values side by side.
If you design social graphics, slides, infographics or memes, build them on a PNG base. The Meme Generator and Watermark Tool both produce cleaner output when the starting image is lossless.
A frequent reason to want PNG is transparency, for example a product photo on no background or a logo for a slide deck. Conversion alone does not create transparency, because the JPG never stored any. Once you have a PNG, you can remove the background with the AI Background Remover, which sends the image to the remove.bg service and returns a transparent cut-out. Save the result as PNG or WebP to keep the alpha channel.
Everything on this page runs on your device. The file is read with the browser's FileReader, drawn onto a hidden canvas and exported as a PNG blob. There is no server round-trip, so conversion is fast even on slow connections, and private images such as IDs, receipts or medical documents never leave your computer or phone. Canvas export also writes a fresh file without EXIF metadata, which removes camera details and GPS coordinates.
Convert client-supplied JPGs to PNG before layering type, shapes and effects, so the final export starts from the cleanest possible base.
Screenshots of slides or diagrams that were saved as JPG stay sharp from now on when annotated as PNG.
Documentation tools and app stores often prefer PNG for UI captures. Convert once and keep the crisp version in your repository.
Many print services want PNG for designs. Convert, then remove the background for die-cut stickers.
No converter can recover detail a JPG has already discarded. What PNG does is lock the image in its current state, so cropping, annotating or re-saving it in an editor causes no further loss. Think of it as stopping the damage rather than repairing it.
PNG stores every pixel exactly, including the noise and the small blocks introduced by JPG compression. Lossless compression cannot shrink that randomness much, so photos typically grow five to ten times. That is expected and is the price of a lossless working file.
Not automatically. A JPG has no transparency information, so the PNG starts fully opaque. To cut out the background, run the PNG through the AI Background Remover (that tool sends the image to remove.bg for processing).
No. Decoding and encoding happen in your browser with the built-in canvas API. Your files never reach our servers, so this is safe for personal and confidential images.
Yes. Drop as many files as you need, convert them together and download the results as one ZIP archive.
Keep the JPG if you only need to view, share or upload the photo. JPG is smaller and universally accepted. Convert to PNG when you are about to edit, when the image contains text or sharp graphics, or when a tool or platform specifically asks for PNG.
The exported PNG does not include EXIF metadata such as camera model or GPS location. Re-encoding through the canvas writes only the image pixels, which is a privacy bonus when sharing.
There is no artificial limit. The practical ceiling is your device memory: very large images (above roughly 50 megapixels) or batches of hundreds of photos are best done on a desktop computer.
Facing issues or want to request a feature?
Let us know immediately! We would love to hear your feedback.