Whisper large-v3 packaged as GGUF runs locally under 1GB without Python
A GGUF-packaged version of OpenAI's Whisper large-v3 built for transcribe.cpp achieves 1.81% WER on LibriSpeech and runs on a laptop in under 1 GB without Python.
Read the full story at AlphaSignal ↗
Timeline · 1 report
- 2026-11-15 16:00 · AlphaSignal
OpenAI's Whisper large-v3 Runs Locally Under 1GB Without Python