Support
Need help with Tesseract? Find answers below or get in touch.
Frequently Asked Questions
What is Tesseract?
+
Tesseract is a personal AI that runs entirely on your Mac. It types what you say into any app, reads selected text aloud, answers in a chat like the AI apps you already know, and takes an Appshot of any window you want to ask about. The Companion, in alpha and in active development, lives the day alongside you.
How is Tesseract different from ChatGPT?
+
ChatGPT runs on OpenAI's servers. Tesseract runs on your Mac's own chip, so there is no account, nothing you type, say, or show it is uploaded, and after the first model download it works without the internet. It also works outside the chat window: it types your dictation into any app and reads any selected text aloud.
What are the system requirements?
+
Tesseract needs macOS 26 or later on an Apple silicon Mac, M1 or later. 16 GB of memory is enough for dictation, voice, and chat, and the Companion wants 48 GB or more. The internet is needed once, to download the models on first launch.
Does Tesseract work offline?
+
Yes. After the one-time model download on first launch, Tesseract works entirely offline. No internet connection is required for any of its features.
Is my data sent to the cloud?
+
No. All processing happens locally on your Mac using Apple silicon. Your voice recordings, conversations, and generated content never leave your device. There are no accounts, no analytics, and no telemetry.
Which languages can I dictate in?
+
Dictation understands 99 languages. It runs on Whisper, entirely on your Mac, so every one of them works offline.
Can other apps use Tesseract's models?
+
Yes. Switch on the server and Tesseract answers on localhost as an OpenAI-compatible API, so coding tools like OpenCode, or any app with a base-URL field, can use the models already on your Mac. It is off until you turn it on, and it only listens on your own machine.
How much disk space do the models need?
+
The initial model download is several gigabytes depending on which features you use. Models are stored locally and only need to be downloaded once.
How do I set up dictation?
+
Launch the app, grant Microphone and Accessibility permissions when prompted, then press and hold Option+Space to dictate. Release to insert the transcription. You can customize the hotkey in Settings.
The app is asking for Accessibility permission. Why?
+
Accessibility permission is needed for two things: registering global hotkeys (so features work from any app) and simulating keyboard paste to insert text into the active text field.
Can I change the hotkeys?
+
Yes, in the app's Settings. The defaults are Option+Space for dictation, fn+Space to read text aloud, Control+Space to talk to the chat, and both Command keys together for an Appshot.
Dictation isn't working. What should I check?
+
Make sure you've granted both Microphone and Accessibility permissions in System Settings > Privacy & Security. Also ensure the model has finished downloading (check the app's status). Try restarting the app if permissions were just granted.
How do I uninstall and remove all data?
+
Delete Tesseract from your Applications folder. To remove all associated data including downloaded models, delete the app's container from ~/Library (or use a tool like AppCleaner).
How do I request a refund?
+
Please reach out via support@thetesseract.app and we'll help you with your purchase.