--- title: Social Image Resizer｜Học viện Tri thức Mai Strategy lang: vi source: https://mindsprt.dev/vi/tools/social-image-resizer/ --- # Social Image Resizer > Turn one key visual into every platform size, with the file staying in your browser Pick an image, choose a platform size and a scaling mode, then download. The three modes behave very differently: Fit keeps the whole image and pads the rest, which suits a key visual that is already laid out; Fill crops the overflow, which suits cut-out product shots or full-bleed backgrounds; Stretch distorts and should only be used deliberately. The size table covers 22 presets across Instagram, Facebook, LinkedIn, X, YouTube, Pinterest, TikTok and web Open Graph, each carrying its source and check date. Everything runs on the browser Canvas, so the file is never uploaded. ## How to Use 1. Click Choose an image to upload a file under 20MB, processed only in your browser and never sent out 2. Pick a Platform and Size, then choose Fit, Fill or Stretch under Scaling 3. With Fit selected you can set the Padding color, and the preview shows the Source and Output size 4. Once the preview looks right, click Download this size to save it as a PNG ## Method & Assumptions The image is redrawn locally on a Canvas and exported as PNG. There is no upload and no outbound request at any point, matching how preflight-checker handles files. Fit scales proportionally until the whole image sits inside the target frame and pads the remainder with your chosen color, cropping nothing. Fill scales proportionally until the frame is covered, centres the image and crops the overflow, with no padding and no distortion. Stretch pulls the image to the target dimensions and distorts it. The interface warns in red, because this is usually not what you want. The size table lives in tools-data/social-sizes.json with a source and check date per entry, and is registered in upstreamSources for periodic re-checking. Platforms change specs routinely, and a table hardcoded into the component would go stale silently. Sizes are the platforms' recommended upload dimensions, which can differ from the actual display crop. Known differences are flagged per entry in the interface, such as Facebook cover art uploading at 851×315 while the desktop display area is around 820×312. ## Use Cases - Catalogue cover visual to a 1080×1080 Instagram square: use Fit with a brand color so the layout is not cropped. - Cut-out product shot to a 1080×1920 story: use Fill so it covers the frame with no letterboxing. - Landscape visual to a 1584×396 LinkedIn banner: an extreme ratio, so check the preview before choosing Fit or Fill. - Switch platform sizes on the same image and download each one, without going back to the design software to re-export. - Produce the 1200×630 Open Graph image in the same pass, so link previews and social assets are done together. - After a post goes live with key art cropped wrong, a social coordinator compares Fit versus Fill to pinpoint which scaling mode was picked by mistake ## FAQ ### Are my images uploaded? No. The file is processed on a Canvas inside your browser with no upload and no outbound request. Close the tab and nothing remains. ### How do I choose between Fit and Fill? It depends on whether the image can be cropped. Use Fit for a laid-out key visual with text or a logo near the edge; use Fill for cut-out shots or full-bleed backgrounds. When in doubt, preview both. ### Why is the Facebook cover 851×315 rather than 820×312? 851×315 is the recommended upload size and 820×312 is the desktop display area. Export at the upload size and keep key elements away from the edges, and both are covered. The interface flags this per entry. ### How often is the size table updated? The source and check date are shown below the tool, and the entry is registered in this site's upstream monitoring for periodic re-checking. Platforms change specs routinely, so if the date looks old, defer to the platform's current guidance. ### Can I export every size at once? One size at a time for now. Batch export would need an extra compression library, and pulling in a dependency for a single feature is not worth it, so the tool stays light. ### What format is the output? Lossless PNG. If you need JPEG compression, run the output through your usual compression tool afterwards. ## Limitations & Disclaimer - Sizes come from a consolidated industry source rather than each platform's own help pages, so they can lag when a platform changes specs. The check date is shown below the tool. - A recommended upload size is not the same as the display crop. Known differences are flagged per entry; verify the rest yourself. - Output is PNG only, with no JPEG quality control and no batch export. - Cropping is always centred, with no custom crop position or focal point. - There is a 20MB limit, so compress very large files first. Large images may be slow on mobile devices. Updated: 2026-08-15 · Method version: v1.0.0 --- > HTML version: https://mindsprt.dev/vi/tools/social-image-resizer/ > MINDS — 麥思印刷整合有限公司 · https://mindsprt.dev