Category: Images

A 3.5 × 4.5 cm passport photo is 413 × 531 pixels at 300 DPI. To make one, crop your photo to a 7:9 shape (slightly taller than wide), resize it to 413 × 531 px, and save it as a JPEG. You can do all three steps free in your browser with ToolBrainy. Your photo is never uploaded.

3.5 × 4.5 cm (35 × 45 mm) is the passport and visa photo size used by the UK, India, the Schengen countries and many others. The US uses a square 2 × 2 inch photo instead. Both are covered below.

Passport photo sizes in pixels

To turn centimetres into pixels, you need a resolution. Photo printing uses 300 DPI (dots per inch), so 1 cm ≈ 118 px.

Size	Where it’s used	Pixels at 300 DPI	Shape
3.5 × 4.5 cm (35 × 45 mm)	UK, India, Schengen/EU and many others	413 × 531 px	7:9
2 × 2 in (51 × 51 mm)	United States	600 × 600 px	1:1
5 × 7 cm (50 × 70 mm)	Canada	591 × 827 px	5:7
3.3 × 4.8 cm (33 × 48 mm)	China	390 × 567 px	11:16

Rules change, and online forms sometimes ask for a different pixel size than the printed photo. Always check the official site for the passport or visa you’re applying for, and follow the form when it disagrees with this table.

Before you start: take a usable photo

• Plain, light background. Stand about half a metre in front of a white or off-white wall.
• Even light, no shadows. Face a window in daylight. Avoid overhead lights that throw shadows under your eyes.
• Camera at eye level, about 1.5 m away. Too close and your face looks distorted. Ask someone else to take it rather than holding the phone out.
• Neutral expression, eyes open, looking straight at the camera. No glasses for most countries, and nothing covering your face.
• Leave space around you. Include your head and the top of your shoulders, with room above your hair. You’ll crop it down.

Step by step: make a 3.5 × 4.5 cm passport photo

Step 1: Crop to a 7:9 shape

Open the Crop Image tool and choose your photo. Drag the corners of the crop box around your head and shoulders. Under the image, the tool shows the crop size in pixels. Aim for a width-to-height ratio of 7:9, for example:

• 700 × 900 px
• 840 × 1080 px
• 1050 × 1350 px

It doesn’t need to be exact to the pixel. A few pixels off won’t be visible after resizing. Most countries want your face (chin to the top of your head) to fill about 70–80% of the photo’s height, so leave a small gap above your hair and keep your face centred.

Step 2: Resize to 413 × 531 pixels

Open the Image Resizer, add the cropped photo, keep Lock aspect ratio ticked and type 413 in Width. If your crop was close to 7:9, the height comes out at about 531. If it’s a few pixels off, untick the lock and type both numbers. The tiny stretch won’t be visible. Set the output format to JPEG and download.

Step 3: Check the file size (for online forms)

A 413 × 531 JPEG is usually well under 100 KB. If the form has a stricter limit, such as 20 KB or 50 KB, run it through the Image Compressor and lower the quality until it fits. Our guide on resizing an image to 20 KB has the exact settings.

US passport photo: 2 × 2 inches

Same steps, different shape:

1. Crop to a square (equal width and height in the crop readout).
2. Resize to 600 × 600 px for a 2 × 2 in print at 300 DPI.
3. For the US online visa form (DS-160), the photo must be a square JPEG between 600 × 600 and 1200 × 1200 px and no larger than 240 KB. 600 × 600 at normal JPEG quality fits easily.

The US also asks for your head (chin to top of hair) to measure 1 to 1⅜ inches, about 50–69% of the photo’s height.

Printing a passport photo at home or at a shop

Pixels only become centimetres when printed at the right resolution. When you print, choose actual size or 100%, never “fit to page”, and make sure the printer or shop prints at 300 DPI. Many photo shops will print several copies on one standard photo print, which is usually cheaper than printing just one.

Common reasons passport photos get rejected

• Shadows on the face or background.
• Head too small or too big in the frame. Crop closer or leave more space.
• Stretched or squashed face from resizing with the wrong ratio. Crop to the right shape first.
• Heavy filters or beauty retouching. Most passport offices reject edited faces.
• Wrong file format. Use JPEG (.jpg) unless the form asks otherwise. Not sure why? See JPG vs PNG vs WebP.

Frequently asked questions

How many pixels is a 3.5 x 4.5 cm photo?

413 × 531 pixels at 300 DPI. At 600 DPI it would be 827 × 1063, but online forms almost always want the 300 DPI size or smaller.

Is 3.5 x 4.5 cm the same as 35 x 45 mm?

Yes. Some forms write it as 4.5 × 3.5 cm (height first). It’s the same photo, taller than it is wide.

Can I make a passport photo on my phone?

Yes. The ToolBrainy crop, resize and compress tools all work in mobile browsers on Android and iPhone, with nothing to install.

Is it safe to edit my passport photo online?

The ToolBrainy image tools process the photo in your browser, so it isn’t uploaded to a server.

Can I change the background colour to white?

The easiest route is to take the photo against a white wall. Editing the background later often leaves visible edges around your hair, which can get the photo rejected.

The short version

Take the photo against a plain wall in daylight → crop to 7:9 in the Crop Image tool → resize to 413 × 531 px in the Image Resizer → save as JPEG and check the size. For the US, crop square and resize to 600 × 600 px.

To get an image under 20 KB, make it smaller in pixels first, then lower the JPEG quality. Resize the photo to about 300 px wide, then compress it to JPEG at around 60–70% quality, and check the size. For 50 KB, around 600 px wide at 70% usually works. You can do both steps free in your browser with ToolBrainy.

Exam registrations, job portals, scholarship forms and government sites love tight upload limits: “photo between 10 and 20 KB”, “signature under 10 KB”, “max 50 KB”. A phone photo is often 2–5 MB, more than 100 times too big. Here’s how to get it down without making it unrecognisable.

Why a quality slider alone isn’t enough

File size depends on two things: how many pixels the image has and how heavily each one is compressed. A 4000 × 3000 phone photo has 12 million pixels. Even at very low quality, that’s too much data for 20 KB, and the result looks blocky.

Cutting the pixels does most of the work. A 300 × 400 image has just 120,000 pixels, 1% of the original, so it fits under 20 KB at a quality where it still looks clean.

Step by step: resize an image to 20 KB

Step 1: Crop to what the form wants

Most forms want your face and shoulders on a plain background, not the whole room. Use the Crop Image tool to cut out everything else. If the form says “3.5 × 4.5 cm”, that’s a 7:9 ratio, slightly taller than wide.

Step 2: Reduce the pixel size

Open the Image Resizer, add your photo and type a width, keeping Lock aspect ratio ticked:

• Under 20 KB: 300 px wide (or the exact pixel size the form gives, such as 200 × 230).
• Under 50 KB: 500–600 px wide.
• Under 100 KB: 800–1000 px wide.

Set the output format to JPEG and download.

Step 3: Compress to the exact limit

Open the Image Compressor, add the resized image, choose JPEG and set Quality to about 70%. The tool shows the new file size:

• Still too big? Lower the quality in steps of 5–10%.
• Below 50% and still too big? Go back to Step 2 and use a smaller width. That looks better than crushing the quality.
• Too small? Some forms have a minimum, like 10 KB. Raise the quality until you’re inside the range.

Step 4: Check it like the examiner will

Open the final image at full size. Is your face clear, with no blocky patches around your eyes and hair? Is the background plain? If so, upload it.

Starting settings by target size

Target	Width	JPEG quality	Typical use
Under 10 KB	140–200 px	60–70%	Signatures, small ID photos
10–20 KB	200–300 px	60–75%	Exam and job form photos
Under 50 KB	500–600 px	65–75%	Profile photos, ID scans
Under 100 KB	800–1000 px	70–80%	Documents, certificates

These are starting points. Busy backgrounds, grain and low light make files bigger, so always go by the size the tool shows.

Signatures: the easiest file to get tiny

Signature uploads often need to be under 10 KB and around 140 × 60 px. Sign in black ink on white paper, photograph it in daylight, crop tightly around the signature, resize to the requested pixels, then compress to JPEG at about 70%. A clean signature usually comes in at just a few kilobytes. If the paper looks grey, the Background Remover can help clean it up first.

Common mistakes

• Uploading a PNG. PNG is great for logos and screenshots, but photos come out much bigger than JPEG. Use JPEG (.jpg) unless the form asks for something else. Not sure why? Read JPG vs PNG vs WebP.
• Using WebP when the form wants JPG. WebP is smaller, but many government portals only accept .jpg or .jpeg.
• Compressing the same file again and again. Each pass loses more detail. Go back to the original and start over with a smaller width.
• Stretching the photo. Typing both width and height with the aspect lock off distorts faces. Crop to the right shape first, then resize.

Frequently asked questions

How do I reduce a photo to 20 KB on my phone?

Open the ToolBrainy tools in your phone’s browser. They work on Android and iPhone with nothing to install. Crop, resize to about 300 px wide, then compress to JPEG at 60–70%.

Will my photo look bad at 20 KB?

Not if you shrink the pixels first. A 300 px photo at 70% quality looks clean at the size forms display it. It only looks bad if you keep the full resolution and push the quality very low.

Is it safe to upload my ID photo to an online tool?

The ToolBrainy image tools process files in your browser, so your photo isn’t uploaded to a server.

What’s the difference between KB and pixels?

Pixels are the image’s dimensions (for example 300 × 400). KB is the file size on disk. Forms often specify both, and you need to meet both.

Can I compress several photos at once?

Yes. The Image Compressor and Image Resizer both accept multiple images and let you download them all together.

The short version

Crop → resize to about 300 px wide (600 px for 50 KB) → compress to JPEG at 60–70% → check the size → upload. Start with the Image Resizer, then finish in the Image Compressor. Need a PDF under a limit instead? See how to compress a PDF to 100 KB.

You publish a page, open it on your phone away from wifi, and watch the pictures crawl in one band at a time. Almost always the cause is the same: the images went up exactly as they came off a camera or a designer’s export. A single untouched photo can weigh more than every other part of the page combined, and it’s usually the easiest thing on the whole site to fix.

A 6 MB photograph, 4000 pixels wide, sitting in a column roughly 700 pixels across. That exact mismatch turns up in more of the uploads we see than any other single problem. Nothing is wrong with the picture itself — it is simply around thirty times larger than the page has any use for, and every visitor pays for the difference.

Why your images are heavier than they need to be

• They’re far bigger than the space they sit in. Browsers scale a 4000-pixel image down to fit a 700-pixel column, but the visitor still downloads all 4000 pixels first.
• The format doesn’t match the content. A photograph saved as PNG can easily be five times the size of the same photo as JPG, with no visible gain.
• Exports keep quality at maximum by default. Most design tools save at 100% quality, well past the point where anyone can see a difference.
• Hidden data rides along. Camera metadata, colour profiles and thumbnails add weight that no visitor will ever see.

The fastest fix: compress in your browser

If you want the file smaller right now, use a browser-based tool. Our free image compressor re-encodes your image at a sensible quality level and strips the data the web doesn’t need, typically cutting 60–80% off the file with nothing you can spot at normal viewing size.

It runs entirely on your own device, so nothing is uploaded to a server. That matters if you’re working with client photos, product shots that haven’t launched yet, or anything else you’d rather not hand to a third party.

Compressing properly: a practical checklist

1. Resize before you compress. This is the step that does the heavy lifting, and most people skip it. Bring the image down to the largest width it will ever actually be shown at using our image resizer — usually 1600–2000 px for a full-width banner, and far less for anything in a column.
2. Crop away what you don’t need. Empty sky and background are still pixels you’re paying to send. A tighter crop often improves the picture as well as the file size.
3. Choose the right format. Photos want JPG or WebP; logos, icons and screenshots want PNG or WebP. Our image converter handles the switch, and there’s more detail in our guide to JPG vs PNG vs WebP.
4. Then compress. Around 75–85% quality is the sweet spot for photographs. Below roughly 60% you start to see it.
5. Keep the original. Compression is one-way. Store one clean master copy so you can start again if you ever need a bigger version.
6. Check it at full size on a real screen. Zoom to 100% and look at faces, fine text and any flat areas of sky or wall — that’s where problems show first.

What to actually aim for

Useful targets rather than rules, for typical website images:

• Full-width hero image: 1600–2000 px wide, under 200 KB.
• In-article or blog image: 1000–1200 px wide, under 100 KB.
• Product thumbnail or card: 400–600 px wide, under 40 KB.
• Logo or icon: exact display size, under 20 KB.
• Whole page: try to keep every image on it adding up to under 1 MB.

If a single image on your site is over 500 KB, it’s almost certainly resizeable rather than un-compressible.

When compression genuinely shows

Some images resist compression, and it’s worth knowing which so you don’t blame the tool. Screenshots containing small text go mushy quickly, because the sharp edges are exactly what lossy compression discards first. Large flat gradients — a clear sky, a plain studio backdrop — can develop visible banding. Illustrations with hard lines suffer in the same way photographs don’t.

The pattern we run into most is people compressing an already-compressed file. Each pass discards a little more, and after three or four rounds a photo looks visibly rough for no extra saving. If an image already looks slightly soft before you start, go back to the original rather than squeezing it again. For anything with text or sharp edges, switch it to PNG or WebP instead of pushing the JPG quality lower.

Questions people ask about compressing images

How much can I compress an image before it looks bad?

For photographs, 75–85% quality is invisible to almost everyone. The visible threshold usually sits somewhere around 60%, and lower for images containing text.

Should I resize or compress?

Both, in that order. Resizing removes pixels you never needed; compressing makes the remaining pixels cheaper to store. Resizing saves far more.

Does compressing images actually help SEO?

Indirectly, yes. Page speed is a ranking signal and image weight is usually the largest single factor in it. The bigger effect is on visitors, who leave slow pages.

Is WebP worth switching to?

For the web, generally yes — typically 25–35% smaller than JPG at matching quality, with support in every current browser. Keep a JPG if the file also has to open in older desktop software.

Will compressing remove my image’s metadata?

Usually, and that’s normally a good thing for the web. If you need to keep copyright or camera data, check the tool’s behaviour and keep an untouched original either way.

Before you upload

Resize first, pick a format that suits the content, compress once, and keep the master copy. Those four habits will take most sites from sluggish to quick without changing a single line of code. When your images are ready, run them through our image compressor and check the saving before you upload.

You have eight photos of a signed form, or a pile of receipts your accountant wants, and the instruction is always the same: send it as one PDF. Attaching eight separate JPGs looks careless, and whoever receives them has to open each one in turn. The conversion itself takes about thirty seconds — it’s the order, the orientation and the file size that quietly cause all the trouble.

The pattern we run into most often is not really a conversion problem at all. Someone converts their images one at a time, ends up with eight separate PDFs, then goes hunting for a merge tool to undo the mess. What they wanted from the start was simply “all of these in one file, in this order, the right way up.” Get that sequence right and the rest takes care of itself.

Why you keep getting asked for a PDF

Images are fine for looking at. PDFs are what organisations ask for, and usually for good reasons:

• One file, one fixed order. Pages stay in sequence. Nobody has to work out whether IMG_4471 comes before or after IMG_4468.
• It prints predictably. A PDF carries a page size, so it comes out of the printer the way you intended rather than filling a sheet at random.
• Upload forms often accept nothing else. Job applications, visa portals, insurance claims and university submissions frequently reject image files outright.
• It opens anywhere. Phone, laptop, an ancient office PC — a PDF looks identical on all of them.

The fastest fix: convert in your browser

If you just need the file now, use a browser-based converter. Our free JPG to PDF converter lets you select several images at once, arrange them into the order you want, and export the lot as a single PDF. There’s no signup and no watermark.

The part that matters most for documents: it runs entirely in your browser, so your files are never uploaded to a server. If you’re converting an ID card, a bank statement or a signed contract, that difference is worth caring about.

Converting cleanly: a practical checklist

1. Sort your files before you start. Most converters add images in the order you select them. Renaming to 01, 02, 03 takes a moment and saves you re-ordering pages later.
2. Rotate first, not after. Sideways pages are the single most common complaint, and they’re far easier to fix in the image than in the finished PDF.
3. Crop out the desk. Photos of documents almost always include the table edge, a shadow or your thumb. A quick pass with our crop tool makes the result look scanned rather than snapped.
4. Shrink oversized photos. A modern phone photo is far more detail than any A4 page needs. Running them through our image compressor first keeps the PDF small without any visible difference.
5. Convert odd formats up front. If your files are HEIC, WebP or PNG, put them through our image converter first so everything behaves consistently.
6. Open the finished PDF before you send it. Check every page is present, upright, and readable at 100% zoom. Ten seconds here prevents a resubmission.

Page size and resolution: what to actually pick

Two settings decide how your PDF looks. Neither needs much thought once you know the rule:

• A4 (210 × 297 mm) — the default nearly everywhere outside North America. Use it for anything official.
• Letter (8.5 × 11 in) — use this if the document is going to the US or Canada.
• Fit to image — the page takes the shape of your photo. Good for artwork and screenshots, wrong for anything that will be printed.
• 150 DPI is plenty for reading on screen and emailing. 300 DPI is only worth it for professional printing.

As a rough guide, a 12-megapixel phone photo (about 4000 × 3000 pixels) already exceeds 300 DPI on an A4 page. You are almost never short of resolution — you’re far more likely to have too much of it.

When it genuinely goes wrong

Three problems account for most bad conversions. Pages come out rotated, because phones store orientation as hidden metadata that some tools ignore. The file lands at 30–40 MB, because ten untouched phone photos were dropped straight in. Or the text is legible on screen but mushy when printed, because the original photo was taken at an angle in poor light.

The pattern we see most often is the last one: the conversion is blamed when the photograph was the real problem. No converter can add detail that the camera didn’t capture. Retaking the photo flat, in daylight, with the document filling the frame will beat any setting you can change afterwards. If the file is simply too big, convert first and then run the result through our PDF compressor — that’s usually enough to clear a 10 MB upload limit.

Questions people ask about converting JPG to PDF

Can I put several JPGs into one PDF?

Yes. Select all the images together, arrange them into the order you want, and export once. You’ll get a single multi-page PDF rather than one file per image.

Does converting to PDF reduce image quality?

Not by itself. A straight conversion embeds your image as it is. Quality only drops if you deliberately apply compression, or if you resize the image down beforehand.

Why is my PDF so much bigger than the photos?

It usually isn’t — it’s the sum of them, plus a little overhead. Ten 4 MB photos make a roughly 40 MB PDF. Compress the images first, or the PDF afterwards.

Does it work with PNG, WebP or iPhone HEIC files?

PNG and WebP generally convert without trouble. HEIC is less widely supported, so convert those to JPG first and the rest of the process is identical.

Is it safe to convert ID documents or bank statements?

With a browser-based tool that processes files on your own device, nothing is uploaded anywhere, which makes it far safer than a service that sends your document to a server. Always check how a tool handles files before trusting it with anything sensitive.

Before you send it

Get the order right, rotate and crop before converting, and pick A4 unless you’re sending to North America. Those three habits cover almost every problem people run into. When your images are ready, drop them into our JPG to PDF converter and you’ll have one tidy document in a few seconds.

To crop a JPEG image online, open the ToolBrainy image cropper, choose your photo, drag the crop box around the area you want to keep, and select Crop Image. Then select Download image to save a separate copy.

Keeping a photo sharp takes a little more care than choosing the right rectangle. A small crop may look blurry when enlarged, and saving a JPEG can introduce compression changes. This guide explains how to avoid unnecessary quality loss and what the tool can—and cannot—do.

How to crop a JPEG image online

1. Start with your original photo. Keep an untouched copy. If possible, use the camera original rather than an image downloaded from a messaging app that may have compressed it.
2. Choose the file. Open the free image cropper and select Choose image, or drag a file into the drop area. JPG/JPEG, PNG and WebP are supported.
3. Frame the subject. Drag the corners of the crop box to remove unwanted edges. Move the selection to keep the part you need. Watch the Crop size readout: it shows the selected width and height in pixels.
4. Apply the crop. Select Crop Image. The result shows its pixel dimensions and approximate file size.
5. Save and inspect the result. Select Download image. A JPEG input is saved as a .jpg file with “-cropped” added to the filename. Open it at 100% zoom and check the detail you care about before using it.

The current tool uses a freehand selection. It does not offer preset aspect-ratio locking or boxes for typing an exact crop width and height. For a precise ratio, use an editor with that control rather than relying on an approximate square.

Does cropping reduce image quality?

Cropping removes the area outside your selection. It reduces the image’s total pixel count without needing to enlarge the remaining area. However, the crop operation and the process of saving the result are separate.

ToolBrainy creates a new image from the selected area. JPEG and WebP outputs are re-encoded, so they are not guaranteed to be pixel-identical to the original. PNG uses lossless compression, but saving an already damaged JPEG as PNG cannot recover missing detail.

For photographs, one carefully chosen crop may give a perfectly usable result. Avoid repeatedly opening, editing and saving successive JPEG copies. Return to your original whenever you need a different crop.

A pixel-size example: when a crop becomes too small

Illustrative example—not a measured tool test: suppose your original photo is 2400 × 1600 pixels. You select an area 1200 × 800 pixels. That crop keeps the same 3:2 shape and contains 960,000 pixels, one quarter of the original pixel count.

If you enlarge that crop to 2400 × 1600, an editor has to create additional pixels. It cannot recover the original detail outside your selection, and the result may look soft. If you keep only a 600-pixel-wide area but need a 1200-pixel-wide output, choose a wider crop or start with a higher-resolution photo.

The practical rule is to check your destination’s required dimensions before cropping. Leave enough pixels for the final output and avoid enlarging a small selection unnecessarily.

Cropping versus resizing

Crop when you want to remove a distracting border, frame a face, or show only one part of a screenshot. Resize when you want the whole image at different pixel dimensions.

You may need both: first choose the composition, then use an image resizer if the destination needs a smaller exact output size. Keep intermediate saves to a minimum, especially with JPEG.

Which file format should you use?

• JPEG/JPG: suitable for photographs where a smaller file matters. JPG and JPEG refer to the same image format. Repeated lossy saves can accumulate visible artifacts.
• PNG: useful for screenshots, text, diagrams and transparency. Its lossless compression avoids introducing new compression artifacts, but photographic files can be large.
• WebP: supports both lossy and lossless encoding. Do not assume every WebP export is lossless; the encoder’s settings matter. This cropper does not provide a lossless WebP switch.

The cropper is designed to preserve these supported input formats. If you need another format, use the image converter. For a deeper comparison, read JPG vs PNG vs WebP.

Common cropping problems

Why does my cropped photo look blurry?

Check whether the viewer is enlarging the result. Compare it at 100% zoom, inspect the original for compression artifacts, and check whether you started from a low-resolution copy. A tighter crop can make existing defects easier to notice. Re-encoding may also introduce changes.

Why is the cropped file larger?

Fewer pixels do not guarantee fewer bytes. The original compression settings, output format and encoder affect file size. Compare both dimensions and file size. If you have an upload limit, try an image compressor after the final crop, then check whether important detail is still readable.

Can I undo a crop after downloading it?

You cannot restore the removed area from the flattened cropped file alone. Use your original photo to make another crop. Downloading a cropped copy does not itself overwrite the source file, but take care when moving or renaming files afterward.

What if my phone photo will not open?

Check its format. A phone photo may be HEIC rather than JPEG. This tool’s file picker supports JPEG, PNG and WebP; export a compatible copy with an application that supports the original format before cropping.

Are my photos uploaded?

The cropper processes the selected image locally in the browser using an image canvas. Its cropping code does not upload your photo. This describes file processing; it does not mean the website makes no network requests to load the page or its resources.

Before you use the finished image

Check the crop’s dimensions, inspect it at its intended display size, and keep your original. If the image contains personal information, confirm that the downloaded file shows only what you intend to share. Then use your cropped copy where you need it.

Editorial note: This guide was revised with AI assistance using the tool’s interface and implementation as references. The numerical example is illustrative, not a reported hands-on test.

JPG, PNG, WebP: you’ve typed those extensions a hundred times without thinking about them. Pick the wrong one, though, and your image ends up blurry, three times bigger than it needs to be, or missing the transparent background you were counting on. Here’s how to tell them apart and choose the right one in a couple of seconds.

JPG (JPEG): best for photographs

JPG uses “lossy” compression, which cleverly throws away detail your eye is unlikely to notice. That makes it excellent for photos and any image with smooth gradients and lots of colours: sunsets, portraits, product shots. File sizes stay small, which keeps pages fast.

Use JPG when: the image is a photograph and doesn’t need transparency.

Avoid JPG when: the image has sharp edges, text, or flat areas of colour (like logos or screenshots); JPG creates blurry “artifacts” around those edges. It also can’t store a transparent background.

PNG: best for graphics, logos, and transparency

PNG uses “lossless” compression, so nothing is thrown away; every pixel is preserved exactly. That makes text crisp and edges clean, which is ideal for logos, icons, screenshots, and diagrams. PNG also supports transparency, so you can place a logo on any background.

Use PNG when: you need transparency, or the image has sharp text and hard edges.

Avoid PNG when: the image is a full-colour photograph; PNG files of photos are often several times larger than the equivalent JPG for no visible benefit.

WebP: the modern all-rounder

WebP is Google’s newer format that supports both lossy and lossless compression and transparency, essentially the strengths of JPG and PNG in one. For the same visual quality, WebP files are typically 25–35% smaller than JPG and much smaller than PNG. Every modern browser now supports it, which is why it’s become the default recommendation for the web.

Use WebP when: you’re publishing images on a website and want the fastest possible load times.

Be careful when: a file needs to open in older desktop software or be handed to someone who expects a “standard” photo; a JPG is still the safest universal choice.

Quick decision guide

• Photograph for the web? → WebP (or JPG as a safe fallback)
• Logo or icon with transparency? → PNG (or WebP)
• Screenshot with text? → PNG
• Emailing a photo to someone? → JPG
• Print or archiving the original? → keep the highest-quality source (often PNG or the camera’s original)

Converting between formats

Picked the wrong format, or received a huge PNG that should’ve been a JPG? You can switch formats in seconds with our free image converter, which handles JPG, PNG and WebP directly in your browser. If a photo is simply too large, run it through the image compressor first to cut the file size before you publish or send it.

Sometimes the answer isn’t another image format at all. If you’re sending scans or photos of paperwork to someone, a document is the better container — our guide to converting JPG to PDF covers getting the page order, orientation and file size right.

A note on file size and page speed

Image weight is one of the biggest factors in how fast a web page loads, and page speed affects both user experience and search rankings. As a rule of thumb: choose WebP, resize the image to the largest size it will actually be displayed at, and compress it. Uploading a 4000-pixel-wide photo to show in a 600-pixel box wastes bandwidth on every visit.

Common format questions

Is WebP always better than JPG?

For the web, usually yes: smaller files at the same quality. The main exception is compatibility with older, non-browser software, where JPG remains the universal standard.

Does converting PNG to JPG lose quality?

Slightly, because JPG is lossy, and you’ll lose any transparency. For photos the difference is negligible; for logos and text, keep the PNG.

Which format is best for transparency?

PNG and WebP both support transparent backgrounds. JPG does not.

So which one should you pick?

It really is that short: JPG for photos, PNG for graphics and transparency, WebP when you want the best of both on the web. Keep those three pairings in your head and you’ll rarely get it wrong. And when you do need to switch a file over, our image converter turns it into a one-click job.