Format Converter
Convert images between JPG, PNG, WebP, GIF, and BMP — right in your browser.
How to Convert Image Formats Without Losing Quality or Transparency
Every time you save an image in the wrong format, something is lost — sometimes pixels, sometimes transparency, sometimes a dozen kilobytes that add up across a page. Choosing the right container for an image is not a matter of personal preference; it is a technical decision that affects file size, rendering speed, and visual fidelity. This guide explains what each major format actually does to your image data, when to use each one, and how to convert between them without introducing artefacts or accidentally flattening a transparent logo into a white square.
Why Format Choice Matters More Than You Might Think
Image formats are compression schemes, and each one trades off something different. JPG sacrifices tiny amounts of detail across the whole image in exchange for dramatically smaller file sizes — it was designed for photographs where the human eye cannot detect that loss. PNG preserves every single pixel exactly as you saved it, and it supports full transparency, which is why it dominates in logos, icons, and screenshots. WebP does both jobs reasonably well at smaller sizes than either JPG or PNG, which is why it has become the default export format for modern web tools. GIF is ancient and limited to 256 colours, but it carries animation, which is why it persists despite those limitations. BMP is an uncompressed bitmap that stores raw pixel data — rarely useful on the web, but occasionally required by older Windows software or specific print workflows.
The choice is consequential. A photograph saved as PNG will be five to ten times larger than the same image as a JPG with 90% quality, often with no visible difference. A logo saved as JPG will develop coloured fringe artefacts around its edges because JPG's compression algorithm struggles with sharp lines and solid colour blocks. A WebP image delivered to an older browser may simply fail to display at all without a fallback in place.
Understanding Lossy Versus Lossless Compression
The most important distinction between formats is whether their compression is lossy or lossless.
Lossless formats — PNG, BMP, GIF — store enough information to reconstruct the original image pixel-for-pixel. When you open a PNG and resave it as another PNG, nothing changes. This is essential for anything that will be edited further, because each re-encode in a lossy format compounds the degradation.
Lossy formats — JPG, and WebP when not set to lossless mode — analyse the image and discard data that is statistically unlikely to be noticed. JPG divides the image into 8×8 pixel blocks and stores a compressed frequency representation of each block. At high quality settings this is nearly invisible. At low quality settings you see blockiness, colour banding, and smearing around sharp edges. The key rule: never re-encode a JPG as another JPG. Always convert from an original PNG or RAW file if you want to change JPG quality settings, because encoding already-compressed data amplifies the artefacts.
The Transparency Problem with JPG and GIF
JPG has no transparency channel at all. When you convert a PNG with a transparent background to JPG, the converter has to decide what colour to use where the transparency was. Most tools default to white, which is usually fine for web content on white backgrounds, but disastrous for dark-mode interfaces or anywhere the background colour is not white.
GIF supports one binary transparency — a single colour in its 256-colour palette can be designated as transparent, and pixels are either fully opaque or fully invisible. There is no in-between. This means GIF cannot represent the soft edges and anti-aliasing that PNG handles fluently. Converting a PNG logo with semi-transparent edge pixels to GIF will produce jagged, staircase edges.
PNG supports a full alpha channel: each pixel independently stores its opacity from 0 (fully transparent) to 255 (fully opaque). This is why PNG is the correct format for logos, icons, overlays, and anything that needs to sit on top of varied backgrounds. WebP also supports full alpha transparency, which gives it an advantage over both JPG and GIF for complex imagery with transparency.
When to Use Each Format
Use JPG for photographs, camera images, and any image with continuous colour variation and no transparency requirement. Quality settings between 80 and 92 percent give excellent results for most purposes. Below 80 percent, artefacts become obvious. Above 95 percent, file sizes grow rapidly with little visible gain.
Use PNG for logos, icons, screenshots, graphics with text, and anything requiring transparency or exact colour reproduction. PNG is also the right choice for images that will be edited further, since it preserves all information without loss.
Use WebP for web delivery where you control the server and can verify browser support. WebP at equivalent visual quality is typically 25 to 35 percent smaller than JPG for photographs and 25 to 50 percent smaller than PNG for graphics. Its lossless mode matches PNG quality at smaller sizes. The main caveat is that Safari required iOS 14 or macOS Big Sur to gain WebP support, so very old devices may still not render it.
Use GIF almost exclusively for simple animations with a limited colour range. For static images, PNG is always superior. For animations, WebP and AVIF support animated versions at far smaller file sizes, though GIF remains the most universally supported animated format.
Use BMP when a specific application, particularly Windows desktop software or older print workflows, explicitly requires it. It stores raw uncompressed pixel data, so files are very large and have no place in web delivery.
What Happens During Conversion
A format converter works by decoding the source file back to raw pixel data in memory — essentially reconstructing the full uncompressed image — and then re-encoding that raw data in the target format. This means a good converter does not transcode from one compressed representation to another; it always passes through the uncompressed stage. The quality of the result depends entirely on what the target format can represent and what settings you apply.
Converting PNG to JPG at high quality is nearly lossless in practice for photographs. Converting JPG back to PNG is lossless from that point forward, but it cannot recover detail that the original JPG compression already discarded. Converting PNG to WebP at lossless mode produces a smaller file with identical pixels. Converting to GIF will reduce the colour palette to 256 colours, which is immediately visible on photographs but barely noticeable on simple graphics that already use few colours.
Quality Settings and When to Adjust Them
Quality settings apply to JPG and WebP only, because PNG, GIF, and BMP are lossless (or fixed-depth in GIF's case). The quality slider controls how aggressively the encoder discards information.
For web photographs, 85 to 92 percent is the standard range. Below 85, compression artefacts become noticeable on large displays. Above 92, the file size climbs with diminishing visual return. For thumbnails that will be displayed small, 75 to 80 percent is often acceptable. For images that will be printed or displayed large without further recompression, 95 percent or higher is safer.
WebP quality at 80 percent typically matches JPG quality at 90 percent visually while producing a smaller file, which is one of its main selling points.
Common Conversion Mistakes to Avoid
The most frequent mistake is converting a JPG to PNG thinking you are upgrading it — you are not recovering any lost detail, you are just storing the already-degraded image in a lossless container at a much larger file size. If you want to improve image quality, you need to go back to the original uncompressed source.
A close second is converting transparent PNGs to JPG without explicitly handling the transparency. Always check what colour will replace the transparent areas and confirm it works against every background the image will appear on.
Third is saving GIF at large dimensions. Because GIF is limited to 256 colours, large photographic images converted to GIF will look posterised and terrible. Keep GIF for small icons, simple line art, and short animations where the colour limitation is not visually damaging.
Converting between lossy formats repeatedly — for example, taking a JPG, editing it slightly, saving as JPG, then doing it again — compounds artefacts with each cycle. If you are doing iterative editing, save intermediate versions as PNG to preserve fidelity, and only produce a JPG as the final export step.
Browser-Based Versus Server-Based Conversion
Converting images directly in the browser using the Canvas API has one significant advantage: your image never leaves your device. For sensitive or confidential images, this matters. The converted result is generated entirely by your local CPU using JavaScript's built-in encoding capabilities.
The tradeoff is that browser encoders are generally good but not specialised. Dedicated tools like ImageMagick, libvips, or Squoosh's WebAssembly encoders can sometimes produce smaller files at equivalent quality through more advanced compression passes. For most everyday conversions, the difference is not meaningful. For bulk processing of hundreds of images or maximum compression efficiency, a server-side tool is worth the extra step.