Whisper Large V3 Turbo

OpenAI

4.5x faster than Whisper Large V3 with minimal quality loss. Decoder reduced from 32 to 4 layers. Most-downloaded Whisper variant (4.6M+ monthly). Best speed/accuracy balance for local STT.

Speech-to-Text Local Latest Whisper Family vv3-turbo

Documentation Back to Models

Type

Speech-to-Text

Source

Local

License

MIT

Whisper Family Timeline 2 versions

Whisper Large V3 Turbo Current Latest local

Oct 2024

Whisper Large v3 Latest api

Capabilities

🌊

Streaming

Local Model Specs

Architecture

Transformer Encoder-Decoder (4-layer decoder)

Runtime

faster-whisper / whisper.cpp

Disk Size

1.6 GB

Details

Release Date: October 1, 2024
Knowledge Cutoff: -
Source: Local
License: MIT
Model ID: whisper-large-v3-turbo-local

Last updated: March 13, 2026