Crop JPG

How to Crop JPG Images Online

1. Choose Your JPG Files: Drag and drop JPG images onto the page or click "Choose JPG files" — up to 50 at once. The first photo opens in the editor and the rest line up as thumbnails below it; with two or more photos open, "Add more" adds files to the same batch.
2. Set the Crop: In Crop, pick an "Aspect ratio" ("Free", "Original", 1:1, 4:3, 16:9 and more) or a platform size, then drag the box on the photo or type its width and height under "Crop area". "Shape" switches between "Rectangle" and "Circle (avatar)". The crop applies to every photo in the batch.
3. Pick the Output Format (Optional): "Save as" keeps "Same as original" or saves as JPG, PNG, WebP or another format. For JPG, WebP and AVIF output, the "Quality" slider (40–100) trades file size against detail. Resize, Crop, Rotate and Flip all work in the same session, and "Undo" steps back.
4. Download: Click the Download button, which names the output format (for example "Download PNG"), for one photo, or "Download all (.zip)" for a batch — "Download one by one" saves the files separately. JPG files are processed right in your browser, so they are not uploaded. No sign-up, no watermark.

Why Crop a JPG?

Cropping is a destructive but cheap edit: you keep a rectangular slice and throw the rest away. For JPG specifically, a crop always means a re-encode (JPG can't store the trimmed area as transparency the way PNG does), so the saved file is slightly recompressed — but at high quality settings the visible loss is negligible compared with the framing benefit. Common reasons to crop a JPG:

• Match a social-media aspect ratio exactly — Instagram feed is 1:1 (1080×1080), Stories and Reels are 9:16 (1080×1920), YouTube thumbnails are 16:9 (1280×720 minimum, 2 MB max), and LinkedIn profile photos are 1:1 (LinkedIn accepts 400×400 minimum but recommends 640×640 or larger before its own compression).
• Prepare a visa or passport photo — US visa applications require a square JPG between 600×600 and 1200×1200 pixels, ≤240 KB, with the compression ratio capped at 20:1; cropping a phone snap to a centred square is the fastest way to hit the pixel numbers (the 240 KB cap is a job for the separate Compress JPG tool).
• Tighten product shots for e-commerce — Amazon, Shopify, and Etsy all want consistent framing across a catalogue; cropping to a fixed 1:1 box removes background clutter and keeps thumbnail grids visually clean.
• Remove scanner borders from document scans — Phone-scan apps and flatbed scanners often leave black bezels or stray fingertips; a quick crop trims them without touching the document content.
• Reframe a portrait for a profile picture — A 12 MP camera shot is usually 4:3 or 3:2; cropping to a tight 1:1 around the face produces an avatar that survives platform downsampling (Slack, GitHub, Discord all crop to a circle, so keep the face centred).
• Cut a banner or header from a wider photo — A 16:9 or 3:1 landscape strip can be lifted out of a 4:3 vacation shot to make a Twitter header (1500×500) or LinkedIn cover (1584×396) without re-shooting.

If your source is a PNG, HEIC, or WebP, use Crop PNG or the general Crop Image tool first. If you only need a different size at the same framing, Resize JPG is lossier-friendly than crop. To recompress the cropped output further (for example, to fit Gmail's 25 MB attachment limit), pass the result through Compress JPG.

Aspect Ratio Cheat Sheet — Common Platforms and Use Cases

Pixel dimensions below are the platform's documented recommended size; the JPG cropper accepts any of these as exact width × height inputs, or you can use the preset ratio and re-size after.

Crop vs Resize vs Rotate — Which Edit Do You Need?

Need rotation instead? Rotate JPG handles 90°/180°/270° and free-angle. To switch formats while keeping the crop, Convert JPG to PNG gives lossless storage with transparency support.

Frequently Asked Questions

Does cropping a JPG reduce image quality?

Cropping discards pixels outside the selected rectangle, which is a one-way edit, but the kept area is re-encoded at the "Quality" slider's setting (40–100; it does not copy the source file's own JPG quality) — a high setting introduces no visible loss for one generation. Avoid repeatedly cropping and re-saving the same JPG dozens of times; each pass adds a small amount of generational JPG noise around hard edges. If you need a pristine archival copy, keep the original and treat the cropped file as a working version.

Can I set exact pixel dimensions, or only aspect ratios?

Both. Type a precise Width and Height for pixel-perfect output — useful when a platform requires an exact size (US visa photos at 600×600, YouTube thumbnails at 1280×720, LinkedIn at 640×640) — or pick a ratio preset, and either way drag the crop box into place; there's no separate X/Y position field, so position always comes from dragging, while the dimensions update live and stay editable so you can fine-tune the numbers before you download. Free mode lets you ignore ratios entirely and select any rectangle.

How do I crop a JPG into a circle for a profile picture?

Under "Shape", pick "Circle (avatar)". JPG cannot store transparency — every JPG is a solid rectangle — so the editor switches "Save as" to PNG and saves a square with see-through corners. If the platform rounds avatars itself (LinkedIn, Slack, GitHub, Discord all display avatars as a circle), a centred 1:1 square saved as JPG is all you need. Crop PNG works the same way for PNG sources.

Can I crop multiple JPG files to the same dimensions at once?

Yes — drop up to 50 JPG files at once (a mixed batch works too). The crop region (or aspect-ratio preset) you set applies to every file in the batch; "Download all (.zip)" saves every cropped result in one ZIP, or use "Download one by one" to save them separately.

Why is the output file size larger (or smaller) than I expected?

Cropping removes pixels, so the raw image data shrinks proportionally — a 4000×3000 photo cropped to 1080×1080 keeps roughly 9% of the pixels. But the final JPG also depends on output quality and the complexity of the kept region. A crop that captures detailed texture (foliage, fabric, fur) compresses less efficiently than a crop of smooth sky or solid background, so two same-pixel-count crops can produce different-sized files. To force a specific size, follow up with Compress JPG and set a target file size in MB.

What's the difference between cropping and resizing — can't I just resize to the same dimensions?

Resizing keeps the entire image but scales it; a 4000×3000 photo resized to 1080×1080 stretches the composition into a square, distorting faces and objects. Cropping keeps the original pixel scale but trims the framing — your subject stays the same shape and size in the final file. For most social-media and ID-photo workflows the correct sequence is: crop first to fix aspect ratio, then resize if the cropped pixel count still doesn't match the target spec.

Does cropping preserve EXIF data and the date the photo was taken?

The cropped JPG is a re-encoded file, but standard EXIF tags (camera model, ISO, shutter speed, date taken, GPS coordinates if present) are copied across in most modern tools, including this one. The orientation tag is reset to normal because the pixels are turned upright. If you want to strip metadata for privacy before sharing — for example, removing GPS coordinates from a phone photo — save as PNG or WebP under "Save as" (EXIF is kept on JPG output only); a cropped JPG still carries the location.

Can I crop a JPG without re-encoding it (true "lossless crop")?

True lossless JPG cropping is only possible when the crop boundaries fall on JPG's 8×8 (or 16×16 for chroma-subsampled images) DCT block grid — a niche operation supported by command-line tools like jpegtran -crop. The xconvert cropper re-encodes at high quality for predictable output, which is what most use cases (social media, ID photos, e-commerce) want. If absolute byte-level fidelity matters and your crop happens to align with block boundaries, jpegtran or exiftran on a local machine is the right tool.