# Whisper large-v3-turbo requirements

Updated: 2026-06-15

809M Speech to text, released 2024-10.

- Peak memory: ~1.5 GB at int8
- Tools: whisper.cpp, faster-whisper, MacWhisper, WhisperKit
- License: MIT (commercial use: yes)
- Runs on 39 of 39 tracked devices

Sources: https://huggingface.co/openai/whisper-large-v3-turbo, https://github.com/SYSTRAN/faster-whisper/issues/1030, https://github.com/ggml-org/whisper.cpp
More: https://localmodel.run/model/whisper-large-v3-turbo