# Hex

> Hex (v3.3.0, Nov 2026, ~4k stars) is a dictation app. Hold a shortcut, speak, and it pastes the text into the app you are in. The first run downloads a speech model and transcription then stays on your device; update checks still use the network. Optional Modes can rewrite text through OpenCode. MIT and free. Macs need Apple silicon and macOS 15; Windows and Linux are betas that need a 2013 or newer AVX2 processor on x86\_64. The project does not take pull requests, only issues. Whether it collects usage data was not established.

- Repository: https://github.com/anomalyco/hex
- Official docs: https://github.com/anomalyco/hex#readme
- Checked: 2026-11-03

## What it replaces
Nothing people pay for: a free tool, listed because people search for it.

## Facts, with sources

| Question | Status | Answer | Source | Checked |
|---|---|---|---|---|
| Runs on my machine? | Confirmed | Yes; transcribes on your device after the first model download | https://github.com/anomalyco/hex#how-it-works | 2026-11-02 |
| Code licence | Confirmed | MIT | https://github.com/anomalyco/hex/blob/dev/LICENSE | 2026-11-02 |
| Model weights licence | Partial | Depends on the model you pick from its catalogue of about 15 (Parakeet, Whisper, Qwen3-ASR, SenseVoice and others); Hex lists only Parakeet Ultra, under CC-BY-4.0, in its notices. The other models' licences are not listed by Hex | https://github.com/anomalyco/hex/blob/dev/THIRD_PARTY_NOTICES.md | 2026-11-03 |
| Commercial use | Partial | The app is MIT, which allows commercial use; the speech model's terms were not checked | https://github.com/anomalyco/hex/blob/dev/LICENSE | 2026-11-02 |
| Paid services needed? | Confirmed | None required; dictation works without OpenCode, which is only used for optional rewriting | https://github.com/anomalyco/hex/blob/dev/docs/features/dictation.md | 2026-11-02 |
| Hardware | Partial | x86\_64 needs a 2013 or newer AVX2 processor; Windows x64 needs a current graphics driver for Vulkan; Macs need Apple silicon; no RAM figures | https://github.com/anomalyco/hex/blob/dev/docs/windows.md | 2026-11-02 |
| Operating systems | Partial | macOS 15 or later on Apple silicon; Windows 10/11 and Linux (x64 and ARM64) are betas | https://github.com/anomalyco/hex/blob/dev/docs/windows.md | 2026-11-02 |
| Setup effort | Confirmed | Easy; download an installer or use Homebrew, AUR or Nix | https://github.com/anomalyco/hex#downloads | 2026-11-02 |

GitHub: 4091 stars, 240 forks, last push 2026-11-03, archived: no.

## Alternatives to Hex
- [VoiceStudio](https://repoquarry.dev/projects/voicestudio.md): Local desktop app for voice cloning, voice design, video dubbing, dictation, transcription and audiobooks, built on OmniVoice and other engines.
- [whisper.cpp](https://repoquarry.dev/projects/whisper-cpp.md): Dependency-free C/C++ port of OpenAI's Whisper speech recognition, running on CPU or GPU across desktop, mobile and the browser.
- [Handy](https://repoquarry.dev/projects/handy.md): Free offline speech-to-text for Windows, macOS and Linux; press a shortcut, speak, and your words appear in any text field.
- [Meetily](https://repoquarry.dev/projects/meetily.md): Local AI meeting assistant that records mic and system audio, transcribes live with Whisper or Parakeet and summarises with an LLM.
- [Voicebox](https://repoquarry.dev/projects/voicebox.md): Local desktop voice studio that clones voices, generates speech with 8 TTS engines, transcribes with Whisper and adds system-wide dictation.
- [faster-whisper](https://repoquarry.dev/projects/faster-whisper.md): Reimplementation of OpenAI Whisper on CTranslate2, up to 4x faster with less memory, with int8 on CPU or GPU.
---
Source: RepoQuarry (https://repoquarry.dev/projects/hex). Data licence: CC BY 4.0. Quotes are from project documentation and belong to their authors.