Social Media Image Resizer
Resize one image to every social platform size at once
Social Media Image Resizer tool
Social Media Image Resizer: key facts
- What it does
- Resize one image to every social platform size at once
- Category
- Image Tools
- Cost
- Free, with no account, sign-up, or install.
- Your data
- Runs entirely in your browser — the files and text you enter are never uploaded to a server.
- Last reviewed
- . Report an incorrect result.
About the Social Media Image Resizer
Every placement on every network has a pixel size it actually wants, and supplying anything else hands the crop decision to an algorithm. A landscape photo dropped into a 4:5 Instagram frame loses its edges; a banner uploaded at the size a page displays rather than the size it stores comes back soft on a sharp laptop screen. This tool holds 94 measured targets across 17 networks — Instagram, Facebook, X, LinkedIn, YouTube, TikTok, Pinterest, Threads, Snapchat, WhatsApp, Discord, Twitch, Reddit, Tumblr, Bluesky, Mastodon and Google Business — plus open-graph and email presets, a custom size with a ratio lock, and a small legacy group for dimensions that have since been superseded.
The point of difference is that you tick many sizes at once instead of converting one at a time. Framing is stored as a focal point inside your own photograph together with a zoom level, not as a crop rectangle belonging to one frame, so a single decision travels correctly across shapes: drag a face to where you want it and the vertical Reel, the square feed post and the wide YouTube thumbnail all keep it, each pulling in its own edges only as far as its own aspect ratio requires. A strip of live thumbnails underneath renders every ticked size through the exact code the exporter uses, so the full output set is visible before a single file is written.
Overlays mark where the interface will sit on top of the artwork — the account row and reply bar on a story, the margins TikTok publishes, the avatar that overlaps a LinkedIn or X banner, the narrower crop a Facebook cover gets on a phone, and the three channel-art regions YouTube reads. Avatar slots draw the circle the platform will mask to. None of it reaches a file; the overlay is a separate pass the exporter cannot call. Fill behaviour runs from crop-to-fill through fit-and-pad, stretch, a blurred enlarged copy of the picture itself, and a flat tone averaged from its own border pixels. Nothing is uploaded — compositing happens on a canvas on your machine, and re-encoding also discards the original EXIF block, GPS coordinates included.
How to use it
- Add your pictures — browse for several at once, drop a batch onto the zone, or press Ctrl/Cmd+V to paste a screenshot straight in.
- Search or scroll the catalogue and tick every placement you want; a platform name, a word like "banner", a size like 1080x1920 or a ratio like 9:16 all narrow the list.
- Drag inside the preview to position the focal point and scroll to zoom, or nudge it with the arrow keys once the preview has focus.
- Leave the safe-zone overlay on to confirm nothing important falls under the story bars, a banner avatar, or a circular avatar mask.
- Pick how the picture meets the frame — crop to fill, pad it out over a colour, a sampled tone, a gradient or a blurred copy of itself, or stretch it.
- Optionally warm it up, drain the colour or push the contrast using the filter presets and adjustment sliders.
- Choose PNG, JPG or WebP and either set a quality level or name a kilobyte ceiling to hit, then export — one tick downloads on its own, several arrive as a ZIP with a folder per network.
Why use this version
- One pass produces the whole set. Two photographs against a dozen placements is twenty-four correctly sized files in a single archive, not twenty-four trips through a converter.
- Because framing is anchored in your picture rather than in a frame, positioning a subject once is enough for portrait, square and landscape output simultaneously.
- Safe-zone and circular-mask overlays catch the mistake that normally only surfaces after publishing — a headline sitting behind a reply bar, or a face clipped by an avatar crop.
- Reductions happen in stages rather than one jump, which is what keeps small output — a 200 px avatar cut from a 24-megapixel original — crisp instead of ragged.
- Naming a kilobyte ceiling lets the tool search for the best quality that still fits, which beats guessing at a slider when something downstream enforces a limit.
- Nothing leaves the machine, so unreleased packaging shots, embargoed client work and family photographs are all safe to run through it.
Frequently Asked Questions
Why is the Facebook cover preset 1640×624 when Facebook shows it at 820×312?
Those are two different numbers. 820×312 is the box a page draws on a standard desktop display; the file behind it is stored larger and served at higher density to anyone on a sharper screen. Hand over only 820 pixels of width and the browser has to invent the rest, which reads as mushy. Supplying double gets the sharp version, and the older figure is still there under Legacy if something in your workflow insists on it.
What exactly do the dashed rectangles on a story or banner mean?
They mark regions where the network draws its own controls over your artwork. On a 1080×1920 story the top strip carries the account row and the bottom strip the reply and link buttons; TikTok publishes its own margins, which the overlay reproduces; and an X or LinkedIn banner has a profile photograph sitting across one corner. Anything you place there is still in the file — it is simply hidden behind chrome once published. The markers live only in the preview and are never composited into an export.
I ticked eleven sizes but only got one file.
A lone output arrives as a plain image, since wrapping one file in an archive only adds a step. Two or more are collected into a ZIP, by default with one folder per network, which you can switch off under Filenames. If the count looks wrong, check the counter above the export button — it multiplies pictures by ticked sizes, so removing a photograph quietly halves the total.
The warning says the result will be upscaled. How much does that matter?
It appears when the frame demands more pixels than the visible slice of your original can supply at the current zoom, and it reports the shortfall as a percentage of one-to-one. Small shortfalls are usually invisible; large ones look plainly soft, and no amount of resampling recovers detail that was never captured. Zooming back out, or starting from a bigger original, is the only real fix — a 2560×1440 channel banner genuinely needs a large photograph behind it.
Which format and quality should I pick?
Photographs are best served by JPG, and around 85–92% is where most people stop being able to tell it from the original. Flat graphics, screenshots and anything needing transparency want PNG. WebP is the smaller modern option where the browser can write it — Safari only learned to encode it in version 16, and the button is hidden where it cannot. Transparency survives in PNG and WebP only, so a transparent background is filled white when the format is JPG, and the preview shows that rather than hiding it.
Does anything get sent to a server?
No. Decoding, framing, filtering, encoding and archive assembly all happen inside the page. There is no account, no queue, and no size cap imposed by an upload. One side effect worth knowing: because each output is re-encoded from a canvas rather than copied, camera metadata does not survive — which quietly strips GPS coordinates out of phone photographs, and also means a copyright or caption field you were relying on is gone.
