NVIDIA · Shipped
Nemotron 3.5 Transcribe release date
Released
Oct 28, 2026
What we know
An open speech recognition model that handles transcription, speaker recognition and translation, driven by text prompts. Nvidia also ships a dataset of over 800,000 hours of speech and fine-tuning tools alongside it.
Spec
- Params
- Not stated
- Active
- Not stated
- Context
- Not stated
- Weights
- Open
- Modality
- Not stated
- Codename
- Not stated
Every claim
2 from 2 sources- TechmemeRepostOct 28
Techmeme relays SiliconANGLE's report that Nvidia released Nemotron 3.5 Transcribe, an open-weight multilingual audio transcription and translation model, today.
- The DecoderPressOct 28
The Decoder reports Nvidia has released Nemotron 3.5 Transcribe, an open speech recognition model, but gives no specific release date.