Fact · Speech recognition
8times the speed of large
The turbo Whisper model runs at about 8 times the relative speed of the large model.
Checked against its source on October 17, 2026
In the source's own words
“| turbo | 809 M | N/A | `turbo` | ~6 GB | ~8x |”
openai/whisper: README (raw text), OpenAI, GitHub (openai/whisper), 2026, Available models and languages table.
Open the original- What it measures
- The README describes turbo as an optimized version of large-v3, and gives the relative speed measured on an A100 GPU.
- What it does not show
- A relative figure from one GPU, for English speech. The README says turbo is not trained for translation.
Source
Cited on
- How to Transcribe Audio to TextGuide
- Video to Text: Transcribe Videos and YouTubeGuide
- How to Run Whisper Locally on Your Own ComputerGuide
- Transcribe Audio With Whisper, ChatGPT or GeminiGuide
- How do you transcribe a long audio file, such as one over 30 minutes?Question
- What is the best GPU for Whisper?Question