Meme Generator

Every caption is its own box

Drop in a picture or an animated GIF — or start from one of the blank templates — and type your captions. Each caption is its own box: drag it on the canvas or nudge it with the arrow keys (hold Shift to move further), add as many as the meme needs, and remove the ones you do not. A picture opens with the familiar pair, one across the top and one along the bottom, and every caption starts in the classic look — uppercase Impact, white fill, thick black outline. Long captions wrap to the width of their own box, and if one ends up drawn outside the picture the page says which one instead of quietly cropping it. Everything happens in your browser, so your image is never uploaded to a server.

Nineteen blank templates

No picture to hand? The library holds 19 blank grounds: eight solid colours and four gradients at 600 × 498, four panel layouts (two-panel, three-panel, a 2 × 2 grid and side-by-side), and the three shapes people post to — 1080 × 1080 square, 1080 × 1920 story and 1920 × 1080 widescreen. Search them by name, by keyword or by pixel size, so typing 1080 narrows the list to the square and the story. A panel template opens with one caption already centred in each panel. Nothing is downloaded for a template — each one is a handful of numbers drawn at export time, so a story template really is exported at 1080 × 1920. They are blank canvases and layouts, not the famous photographs: those are hosted under licence elsewhere, so bring your own picture for one of those.

Styling each caption

Select a caption and it gets its own controls: typeface — Impact, Arial, Verdana, Courier New, Comic Sans MS, Times New Roman, Georgia, and a Hangul gothic stack, because Impact has no Hangul glyphs at all — plus size, horizontal and vertical alignment, bold, italic, uppercase, text colour, outline colour, outline width, a drop shadow drawn in the outline colour, opacity, and the width and height of the caption's own box. The typefaces are font stacks, not downloads: nothing is fetched, so a caption appears the moment it is typed and the tool still works offline. Positions and sizes are kept as fractions of the finished picture, so what you drag on the preview lands in exactly the same place in the file you save, whatever size it is exported at.

The picture, and text outside it

A loaded picture can be rotated a quarter turn either way, flipped across or down, and turned greyscale, sepia or blurred, with one button to put all of it back. If you would rather the words sat off the picture instead of over it, the caption bar adds a band above it, below it or both, with a height from 5% to 60% and a colour of your choosing. Because captions are positioned against the whole canvas, a caption at the top drops into the bar as soon as the bar exists.

PNG, JPG, or a GIF that still moves

Save as PNG, JPG or GIF. An animated GIF is decoded in the page — this tool carries its own GIF reader, down to the LZW bitstream, interlacing and all four disposal methods — every frame is captioned, and it is written back with each frame's own delay and its loop flag, so the result still animates instead of collapsing to a still. Use the original image resolution is on by default, so a photo saves at its own pixel size; turn it off to export 600px wide. Past about 16 million pixels a canvas comes back blank on iOS Safari, so the export is scaled down and the page says so rather than handing you an empty file.

Why make memes locally

Most online meme makers upload your image to their servers, stamp on a watermark, or gate the download behind a sign-up. Here the whole thing runs on your device: no watermark, no account, no upload, and it keeps working offline. The dashed box around the selected caption is a handle for you and is never drawn on what you download. Pictures up to 40 MB are taken by drop, by the file picker, or by pasting one straight from the clipboard.