whisper-base vs whisper-tiny

Specs, VRAM requirements and download trends — updated 6 August 2026.

openai · A
Params
70 M
Context
30d
4.8 M
apache-2.0 · commercial OK
openai · B
Params
40 M
Context
30d
2.0 M
apache-2.0 · commercial OK

Specification comparison

Differences are highlighted; identical values are muted.

Specification whisper-base whisper-tiny
Parameters 70 M 40 M
Architecture WhisperForConditionalGeneration WhisperForConditionalGeneration
Context length
Licence apache-2.0 apache-2.0
Commercial use Allowed Allowed
Languages 99 99
Downloads 30d 4,759,784 2,026,008
Downloads all time 46.7 M 20.3 M
Quantizations on the Hub SAFETENSORS SAFETENSORS
Gated No No
Common Voice 11.0 (reported) 131 141
LibriSpeech (clean) (reported) 5.0087691176193 7.54
LibriSpeech (other) (reported) 12.849362732121 17.15
First seen on the Hub 2022-09-26 2022-09-26

Download trend

Daily snapshots, last 10 days (28 Jul – 6 Aug)

6.4 M1.8 M
whisper-base whisper-tiny

VRAM side by side

On RTX 4090 · 24 GB · 8K context unless noted

whisper-tiny · fp16 0.7 GB / 24 GB
✅ Runs comfortably
whisper-base · fp16 0.8 GB / 24 GB
✅ Runs comfortably

Adjacent comparisons