speaker-diarization-3.1 vs whisper-medium

Specs, VRAM requirements and download trends — updated 27 August 2026.

pyannote · A

Params

—

Context

—

30d

8.0 M

mit · commercial OK

openai · B

Params

760 M

Context

—

30d

378 K

apache-2.0 · commercial OK

Specification comparison

Differences are highlighted; identical values are muted.

Specification	speaker-diarization-3.1	whisper-medium
Parameters	—	760 M
Architecture	—	WhisperForConditionalGeneration
Context length	—	—
Licence	mit	apache-2.0
Commercial use	Allowed	Allowed
Languages	—	99
Downloads 30d	7,950,173	378,075
Downloads all time	347.7 M	20.0 M
Quantizations on the Hub	—	SAFETENSORS
Gated	Yes — accept licence	No
First seen on the Hub	2023-11-16	2022-09-26

Daily snapshots, last 56 days (28 Jul – 21 Sep)

10.0 M342 K

speaker-diarization-3.1 whisper-medium

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

whisper-medium · fp16 4.0 GB / 24 GB

✅ Runs comfortably