Fact · Speech recognition
1GB of VRAM
The base Whisper model needs about 1 GB of video memory.
Checked against its source on October 17, 2026
In the source's own words
“| base | 74 M | `base.en` | `base` | ~1 GB | ~7x |”
openai/whisper: README (raw text), OpenAI, GitHub (openai/whisper), 2026, Available models and languages table.
Open the original- What it measures
- The table lists the model's approximate memory requirement and inference speed relative to the large model. The speed figure is a separate fact.
- What it does not show
- The README's figures are approximate and were measured on an A100 GPU. Real use depends on the audio and the hardware.