Nemotron 3 Diarization
NVIDIA · · 43 days after Nemotron 3.5 Lightning
Verified
1 of 1 published values traced to a source2 primary sources, 2 archived
Why this status
Newly added. Sources are not archived and no value has been traced to one yet — run `npm run enrich`.
What these states mean · How records are judged · Has this record changed?
An open-weights speaker diarization model of about 100M parameters: from 16 kHz audio it reports who spoke when, for up to eight speakers including overlapping speech, in streaming or offline mode. Its Hugging Face repository was created on September 1; NVIDIA announced it on September 23.
Details
| Released | September 23, 2026stated in huggingface.co, huggingface.coNemotron 3 Diarization was released on September 23, 2026.
|
|---|---|
| Company | NVIDIA |
| Family | Nemotron Diarization |
| Era | Agentic systems |
| Type | Audio |
| Modalities | Not recorded |
| Capabilities | Not recorded |
| Weights | Open weights |
| Licence | OpenMDW License 1.1 |
Current information, elsewhere
This record is a historical one: every figure above is traced to a dated primary source. For what is true today — live pricing, current rankings, availability — these are better maintained than anything here could be.
- Current price and providers OpenRouter ↗
- Current benchmark scores Artificial Analysis ↗
- Weights and model card Hugging Face ↗
Sources
Record status: verified · confidence 90/100 · 2 of 2 primary
Newly added. Sources are not archived and no value has been traced to one yet — run `npm run enrich`.
- huggingface.co Official announcement archived
- huggingface.co Model card archived
Primary means published by the organisation that made the model. Verified records require at least one.
Nemotron Diarization family
- Nemotron 3 Diarization September 23, 2026
Compare Nemotron 3 Diarization with another model →
See Nemotron 3 Diarization on the interactive timeline →