whisper.cpp
Whisper in plain C/C++ — fast on CPUs, Apple Silicon and mobile.
- Stars
- 54.4K
- Forks
- 6.3K
- Open issues
- 353
- Last commit
- 2026-10-12
Live from the GitHub API · updated 2026-10-15
What is whisper.cpp?
whisper.cpp is a dependency-free port of Whisper built on the ggml library. It runs efficiently on CPUs and Apple Silicon (Metal/Core ML), supports quantized models and powers many desktop and mobile transcription apps.
Key features
- No Python required
- Metal, CUDA, Vulkan and Core ML acceleration
- Quantized models for low memory use
- Real-time streaming example
System requirements
- Platforms
- Windows · macOS · Linux
- RAM
- ~1–4 GB depending on model
How to install whisper.cpp
Summarized from the project’s README. Projects change quickly — check the GitHub page for the latest instructions.
Build from source
- 1Clone and download a model.
git clone https://github.com/ggml-org/whisper.cpp.git cd whisper.cpp sh ./models/download-ggml-model.sh base.en - 2Build and transcribe a sample.
cmake -B build cmake --build build -j --config Release ./build/bin/whisper-cli -f samples/jfk.wav
Strengths
- Very fast on Apple Silicon
- Easy to embed
Limitations
- Command-line only
- Expects 16 kHz WAV input
whisper.cpp alternatives compared
| Software | Stars | License | Platforms | VRAM |
|---|---|---|---|---|
| whisper.cpp Whisper in plain C/C++ — fast on CPUs, Apple Silicon and mobile. | 54.4K | MIT | — | |
| Whisper OpenAI’s open-source speech recognition and translation model. | 110.5K | MIT | ~1 GB (tiny) to ~10 GB (large) | |
| faster-whisper Whisper reimplemented on CTranslate2 — up to 4× faster with less memory. | 25.9K | MIT | — | |
| WhisperX Fast transcription with word-level timestamps and speaker diarization. | 24.6K | BSD-2-Clause | — | |
| Coqui TTS Deep-learning text-to-speech toolkit, including the XTTS voice cloning model. | 46.1K | MPL-2.0 | — | |
| GPT-SoVITS Few-shot voice cloning and TTS with a full training WebUI. | 62.8K | MIT | 6 GB+ |
FAQ
Is whisper.cpp free?
Yes — whisper.cpp is open-source software released under the MIT license. Check the license (and any model weight licenses) before commercial use.
Does whisper.cpp run on Windows, Mac and Linux?
whisper.cpp supports: Windows, macOS, Linux.
What hardware do I need for whisper.cpp?
RAM: ~1–4 GB depending on model
How do I install whisper.cpp?
1. Clone and download a model. 2. Build and transcribe a sample.
What are the best whisper.cpp alternatives?
Popular alternatives include Whisper, faster-whisper, WhisperX, Coqui TTS. See the comparison table on this page.