whisper-base vs VibeVoice-ASR
Specs, VRAM requirements and download trends — updated 28 July 2026.
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | whisper-base | VibeVoice-ASR |
|---|---|---|
| Parameters | 70 M | 8.7 B |
| Architecture | WhisperForConditionalGeneration | VibeVoiceForASRTraining |
| Context length | — | — |
| Licence | apache-2.0 | mit |
| Commercial use | Allowed | Allowed |
| Languages | 99 | 51 |
| Downloads 30d | 1,544,243 | 738,250 |
| Downloads all time | 49.4 M | 5.4 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2022-09-26 | 2026-01-21 |
Download trend
Daily snapshots, last 56 days (28 Jul – 21 Sep)
6.4 M631 K
whisper-base
VibeVoice-ASR
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
whisper-base · fp16
0.8 GB / 24 GB
VibeVoice-ASR · fp16
20.9 GB / 24 GB