Skip to main content
HyperWhisper runs natively on macOS, Windows, and Linux. Select your platform below. Each tab gives the minimum version, the supported architectures, and the hardware notes for local models.

​
Minimum requirements

  • macOS 14.6 (Sonoma) or later
  • Apple Silicon (M-series) or Intel — universal binary
  • ~200 MB free disk for the app, plus space for each local model you download (78 MB to 3.1 GB each)

​
Apple Silicon vs Intel

On Apple Silicon, local models run faster. The small Whisper model transcribes in realtime, even on an M1 Air.On Intel Macs, local transcription models work, but they use the CPU only. On Intel, use a small local model (Whisper Tiny or Base), or use HyperWhisper Cloud. Local Gemma post-processing needs Apple Silicon. Intel Macs must use a cloud post-processing provider.

​
Network

  • HyperWhisper Cloud and BYOK providers need an internet connection.
  • Local models transcribe offline after you download them.
  • License activation needs one network call.
If you are behind a corporate firewall, see Network Filtering.