Fact · Speech recognition
4times the speed of large
The small Whisper model runs at about 4 times the relative speed of the large model.
Checked against its source on October 17, 2026
In the source's own words
“| small | 244 M | `small.en` | `small` | ~2 GB | ~4x |”
openai/whisper: README (raw text), OpenAI, GitHub (openai/whisper), 2026, Available models and languages table.
Open the original- What it measures
- The speed is relative to the large model, measured on an A100 GPU for English speech.
- What it does not show
- A relative figure from one GPU. It is not a time for any particular recording.