Fake Tweet Generator
Build a realistic X post mockup for a slide, a lesson, or a design comp — then export it as a PNG.
For parody, education, and mockups only
With or without the @.
Engagement
This is a generated image for entertainment purposes.
Everything here runs on your device. Nothing you enter is uploaded or stored.
Type a name, a handle and some text, and this draws a post that looks like the real thing. Every number is editable, the verified badge toggles, and you can switch between the light and dark colour schemes before exporting a PNG.
How to use it
- Enter a display name and a handle. Leave the photo blank and an avatar is generated from the name, or upload your own.
- Write the body text. Mentions, hashtags and links pick up the accent colour automatically.
- Set the reply, repost, like, view and bookmark counts. They abbreviate as you type.
- Optionally attach an image — it is cropped to fill and rounded to match the real client.
- Pick light or dark, then Download PNG or Copy image.
The stat row is what gives a mockup away
Most fake post generators get the layout roughly right and the numbers completely wrong, and the numbers are what people actually notice.
X has specific rules for abbreviating counts. Below a thousand it shows the exact figure. From 1,000 to 99,999 it shows one decimal place — 1.2K, 19.7K — and drops the decimal when it would be a zero, so 2,000 renders as 2K rather than 2.0K. Above 100,000 it stops using decimals entirely and shows whole thousands: 240K. Past a million the same pattern repeats with M.
Views are handled differently again. They sit on their own line next to the timestamp rather than in the icon row, they use the same abbreviation rules, and the word “Views” is spelled out. A mockup that puts views inline with likes, or that writes 1,240,000 in full, is wrong in a way that is immediately visible to anyone who uses the platform daily.
The order of the icon row matters too: reply, repost, like, view, bookmark, left to right. That order has been stable for years, and a shuffled row looks off even to someone who could not tell you why.
This tool implements all of it, which is why the numbers you type come out looking like numbers the platform would render.
Light and dark are different designs
The dark variant is not the light one inverted. The real client uses pure black for the dark background — not a dark grey — because it was designed for OLED screens where black pixels are switched off entirely. Text drops to a warm off-white rather than pure white, the divider lines lift to a mid grey, and the accent blue stays exactly the same in both themes.
Reproducing that properly matters if the mockup is going into a design review, because a dark mode screenshot with a #1a1a1a background is a detail a designer will catch.
What it does not do
It draws a single post. Reply threads, quote posts and multi-image grids are not implemented here. It does not fetch anything from the real platform, so it cannot import a genuine post or verify that a handle exists. Fonts are the closest available web equivalents rather than the platform’s proprietary faces, which is deliberate — licensing those would not be legitimate.
Nothing is uploaded. The canvas is drawn in your browser and the PNG is produced on your device.
Count abbreviation rules
| Actual number | Displayed as |
|---|---|
| 847 | 847 |
| 1,200 | 1.2K |
| 2,000 | 2K |
| 19,700 | 19.7K |
| 240,000 | 240K |
| 1,240,000 | 1.2M |
| 15,000,000 | 15M |
Colours used, by theme
| Element | Light | Dark |
|---|---|---|
| Background | #ffffff | #000000 |
| Primary text | #0f1419 | #e7e9ea |
| Muted text | #536471 | #71767b |
| Divider | #eff3f4 | #2f3336 |
| Accent | #1d9bf0 | #1d9bf0 |
Questions
Is this legal to use?
For parody, education, design mockups and satire, yes. Using it to impersonate someone, spread misinformation, defraud, or harass is not, and may be a criminal offence depending on where you are. The output is a generated image, not a real post.
Why does my export look sharper than the preview?
Because it renders at twice the display resolution. A mockup usually ends up in a slide deck or a case study where it gets scaled up, and a 1x export goes visibly soft as soon as anyone zooms.
Will the numbers look convincing?
They follow the platform's own abbreviation rules — one decimal below 100K, whole thousands above it, and views shown separately. Those rules are the detail most generators get wrong, and getting them wrong is what makes a mockup read as fake immediately.
Does anything get uploaded?
No. The image is drawn on a canvas in your browser and the export is generated on your device. Profile photos you select never leave your machine, and nothing is stored on a server because there is no server.
Last updated