NVIDIALaunchAI ModelsOctober 28, 2026

NVIDIA releases Nemotron 3.5 Transcribe, open 30B speech model

Read original source →huggingface.co

The open-weight MoE model has ~4B active parameters and handles 40 languages, speaker attribution for up to 8 speakers, and translation to and from 31 languages in one checkpoint. It tops Hugging Face's Open ASR Leaderboard for open models at 4.71% mean WER on 9 English sets, but works only on recordings, not live audio.

How this story unfolded

1 day · 5 reports · 5 community posts · 10 of 11 shown

  1. Oct 28
  2. Oct 29

More stories today

Open the live feed