Model · OpenAI · Speech-To-Text
Whisper large-v3 Turbo
OpenAI's distilled Whisper variant. ~8× faster than large-v3 with most of the accuracy retained — the practical default for high-throughput STT pipelines.
- Modality
- Speech-To-Text
- License
- MIT (Open)
- Parameter size
- 809M
- Released
- October 1, 2024
- Last verified
- June 8, 2026
- Runs locally
- Yes
What you need to run it
- Mac
- Yes. Any Apple Silicon Mac (M1 or newer) with at least 8 GB (16 GB recommended) of memory. Newer chips are faster, but memory is what decides whether it runs.
- Windows / Linux PC
- NVIDIA works best. A graphics card with at least 6 GB (8 GB recommended) of video memory.
- No graphics card
- Works. Runs on a normal computer's processor, just slower.
The best size-for-speed Whisper for a Mac, via whisper.cpp or MLX Whisper; usable on a plain CPU, just slower than real time on old machines.
Numbers are for: official fp16 weights (809M). Based on the maker’s own figures; where the maker gives only a graphics-card figure, the Mac memory is our estimate.
Strengths
- Roughly 8× faster than large-v3
- Retains most of large-v3's multilingual accuracy
- MIT-licensed weights
Weaknesses
- Slight quality regression on the hardest non-English audio
- No speaker diarization
Try it
| Where | Type | Notes |
|---|---|---|
| Hugging Face | weights | Open weights, MIT |
| Replicate | hosted | API key required |
| Groq | hosted-api | Free tier |
| whisper.cpp | local | C/C++ port; runs on CPU and Apple Silicon |
Used in solutions
Version history
- Whisper large-v3 Turbo Oct 2024 Current
- Whisper large-v3 Nov 2023
- Whisper large-v2 Dec 2022 Deprecated
Official sources
- Model card model
Change log
- — Added hardware requirements (Mac support, memory, NVIDIA). Source: https://github.com/openai/whisper (about 6 GB VRAM for turbo) + mlx-community/whisper-large-v3-turbo.
- — Initial entry.
Esc