Fuck You Flow
ΕΛ

Open model. Public code. Somebody charged you for it anyway.

free foreverWindowsopen source

The free Wispr Flow alternative for Windows

Fuck You Flow is a free Wispr Flow alternative for Windows: no subscription, no account, no word limit. It speaks Greek and English, and the recognition runs on your own PC, so your voice never leaves the machine. Wispr Flow hands you 2,000 words a week for free and then asks for 12 to 15 dollars a month. This page puts the two side by side, with every number sourced.

Windows 10 and 11. Version 0.9.0 beta. 1.4 GB with every model inside. No internet connection after setup. MIT license. The SHA256 checksum is on the release page.

Fuck You Flow
The Fuck You Flow home screen: the large-v3 engine on the GPU, the right Alt key for start and stop, words dictated today and the last five transcripts in Greek and English.

Side by side

Every row checked by hand. The prices and the weekly word count come from the Wispr Flow pricing page, read on 6 September 2026. The valuation is from TechCrunch, 17 August 2026.

Wispr FlowFuck You Flow
PriceFree to 2,000 words a week, then 12 to 15 dollars a month0, forever
Words per week2,000 on the free planAs many as you can say
Where your voice goesTheir serversInto the text. Then nowhere.
AccountRequiredNone
InternetAlwaysOnly to download it
Operating systemsWindows, macOS, iOS, AndroidWindows 10 and 11
GreekOne of 100+ languagesFirst language, mixed with English in the same sentence
Source codeSecretOpen, MIT, on GitHub
Company valuation2 billion dollars (August 2026)One domain name

Wispr Flow raised 361 million dollars, reported by TechCrunch on 17 August 2026. The whole of Fuck You Flow is a 1.4 GB installer and an MIT license, written in Thessaloniki.

Where Wispr Flow wins

Three things Wispr Flow does that this app does not do today. If one of them is what you need, theirs is the tool that does the job.

It runs on Macs and phones too

Wispr Flow works on macOS, iOS and Android as well as Windows. Fuck You Flow runs on Windows 10 and Windows 11.

It sits in your meetings

Wispr Flow has a meeting notetaker built in.

It needs no graphics card

The work happens on their servers, so a thin laptop keeps up with a gaming PC.

How to switch

Three steps. About ten minutes, most of it the download.

  1. Download the installer.

    One file, 1.4 GB, with every speech model already inside it. Windows shows a blue warning during install, because the installer carries no purchased signing certificate. Click More info, then Run anyway. The SHA256 checksum is on the release page.

  2. Run it once.

    The first start reads your graphics card and picks the speech model that fits it. NVIDIA, AMD and Intel all run through Vulkan, so any of the three will do.

  3. Press right Alt and speak.

    Press it once to start, speak, press it again. The text lands where your cursor was, in any program: Word, Chrome, Slack, Outlook, even a terminal.

There is no account to create and no card to enter. On an RTX 3070 with the large model, the measured wait from key release to text on screen was 0.17 to 0.88 seconds.

Questions

Is Fuck You Flow really free?

Yes. It costs zero and there is no paid tier, no trial and no card. The code is open under the MIT license on GitHub, so anyone can read it, build it and keep it. The download is one installer of 1.4 GB with the speech models already inside.

Is there a word limit?

None. Wispr Flow gives 2,000 words a week on its free desktop plan, read from wisprflow.ai/pricing on 6 September 2026. Fuck You Flow counts nothing and stops nothing. You can dictate as many words as you can say.

Does it work offline?

Yes. The internet is needed once, to download the installer. After that everything happens on your own machine. Recognition runs on your graphics card through Vulkan, so NVIDIA, AMD and Intel cards all work. Without a graphics card it runs on the processor, more slowly.

Does it work on Mac?

Not today. Version 0.9.0 beta runs on Windows 10 and Windows 11 only. Wispr Flow covers macOS, iOS and Android as well, so on those machines it is the tool that does the job.

Is my dictation private?

Your voice becomes text on your own PC and then it is gone. There is no server, no account and no analytics. Password fields are detected and left empty. The source code is open, so you can check all of this yourself.