| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| latest.json | 2026-10-11 | 13.0 kB | |
| SpeakoFlow_2.0.1_x64-setup.exe | 2026-10-11 | 40.0 MB | |
| SpeakoFlow_2.0.1_x64_en-US.msi | 2026-10-11 | 86.6 MB | |
| SpeakoFlow_x64.app.tar.gz | 2026-10-11 | 37.6 MB | |
| SpeakoFlow_2.0.1_x64.dmg | 2026-10-11 | 38.7 MB | |
| SpeakoFlow_2.0.1_amd64.AppImage | 2026-10-11 | 154.1 MB | |
| SpeakoFlow_2.0.1_amd64.deb | 2026-10-11 | 42.5 MB | |
| SpeakoFlow_2.0.1_aarch64.AppImage | 2026-10-11 | 150.6 MB | |
| SpeakoFlow_2.0.1_arm64.deb | 2026-10-11 | 44.4 MB | |
| SpeakoFlow_aarch64.app.tar.gz | 2026-10-11 | 34.2 MB | |
| SpeakoFlow_2.0.1_aarch64.dmg | 2026-10-11 | 35.1 MB | |
| README.md | 2026-10-11 | 3.6 kB | |
| SpeakoFlow 2.0.1 source code.tar.gz | 2026-10-11 | 39.9 MB | |
| SpeakoFlow 2.0.1 source code.zip | 2026-10-11 | 40.4 MB | |
| Totals: 14 Items | 744.2 MB | 2 | |
SpeakoFlow 2.0.1
A smaller update to 2.0, with a connection test for AI cleanup, a new Thinking setting, and a round of fixes.
On 2.0.0? SpeakoFlow offers this update by itself, or you can check in Settings, About. On 1.4 or earlier? Download 2.0.1 below and install it over your current version. Your settings, history, and models are kept.
New
- Test connection for AI cleanup. The AI cleanup card in Models now has a Test connection button, like the Assistant's. It checks the provider and model that clean up your dictations, and tells you what to fix if something is missing.
- Thinking. Choose how much a reasoning model thinks before it answers (Off, Low, Medium, or High), separately for AI cleanup and the assistant. Off is the default and works as before.
Fixed
- Writing styles now change the text. Formal, Concise, and the other styles used to make no difference.
- Settings no longer reset after a restart or shutdown. Saves can't damage the settings file anymore, and if it is ever damaged anyway, your settings come back from a backup.
- On-device AI no longer crashes loading Gemma 4 on computers with two graphics cards, or with the model only partly on the graphics card. The first time you use on-device AI after updating, a newer engine downloads once in the background. If you're offline, the current one keeps working.
- AI cleanup with gpt-oss no longer pastes the transcript unchanged.
- Text you never said no longer appears after short dictations with the built-in engine.
- macOS: the menu bar shows the SpeakoFlow logo instead of a plain square.
- Linux on KDE Wayland: the assistant can see the text you selected, and its panel responds to clicks again.
Read the full [changelog](https://github.com/AbhishekBarali/SpeakoFlow/blob/main/CHANGELOG.md#201
Which file do I download
| Your computer | Download this |
|---|---|
| Windows | SpeakoFlow_2.0.1_x64-setup.exe |
| Windows, installed by IT or in bulk | SpeakoFlow_2.0.1_x64_en-US.msi |
| Mac with Apple silicon (M1 and newer) | SpeakoFlow_2.0.1_aarch64.dmg |
| Mac with Intel | SpeakoFlow_2.0.1_x64.dmg |
| Linux: Ubuntu, Debian, Mint, Pop!_OS | SpeakoFlow_2.0.1_amd64.deb |
| Linux: any other distribution | SpeakoFlow_2.0.1_amd64.AppImage |
| Linux on ARM: Ubuntu, Debian | SpeakoFlow_2.0.1_arm64.deb |
| Linux on ARM: any other distribution | SpeakoFlow_2.0.1_aarch64.AppImage |
The other three files are not for you to download. Installed copies of SpeakoFlow fetch them by themselves to update: latest.json tells the app a new version exists, and SpeakoFlow_aarch64.app.tar.gz / SpeakoFlow_x64.app.tar.gz are the update packages for Apple silicon and Intel Macs.
Windows: if SmartScreen warns you, choose More info, then Run anyway.
macOS: if it says the app "is damaged", run this once in Terminal (why?):
:::bash
xattr -dr com.apple.quarantine /Applications/SpeakoFlow.app
