faster-whisper
Top pickWhisper, several times faster, on less memory. The practical default.
95
sovereigntyfaster-whisper reimplements Whisper inference on CTranslate2, delivering roughly a four-fold speedup over the reference implementation with substantially lower memory use, and the same transcription output — these are the same weights, executed better. It supports int8 and float16 quantisation, batching, and word-level timestamps, and runs on both GPU and CPU. For most teams replacing a paid transcription API, this is simply the correct starting point.
OPEN SOURCEMITSELF-HOSTLOCAL-FIRST
LicenseMIT
PricingFree. Hardware you already own; a laptop handles the smaller models.
Open sourceYes
Self-hostableYes
Local-first dataYes
What it does well
- +Several times faster than reference Whisper at equal accuracy
- +Quantisation options let large models fit modest GPUs
- +Runs on CPU when no GPU is available
- +MIT licensed, no per-minute cost, nothing leaves your machine
Where it falls short
- −Batch-oriented; streaming needs extra work to do well
- −Diarization is not included — pair with WhisperX or pyannote
- −Accuracy varies by language more than the managed services do
faster-whisper as an alternative to
Where faster-whisper shows up in our comparisons, and how it ranked.
faster-whisper head-to-head
Straight comparisons against the tools people weigh it against.