Nvidia Releases Open-Weight Nemotron 3.5 Transcribe for Recorded Audio

Nvidia Releases Open-Weight Nemotron 3.5 Transcribe for Recorded Audio

Nvidia has released Nemotron 3.5 Transcribe, a free open-weight model, meaning its weights are publicly downloadable. The 30-billion-parameter model runs on a single chip. It transcribes up to 40 languages and handles recordings up to one hour, distinguishing up to eight speakers per file. It does not process live streams. TechCrunch reported the release on October 28, 2026.

Published