Provider model identifier not recorded · Released 11 August 2026
| Provider | NVIDIA |
|---|---|
| Family | NVIDIA Nemotron 3 |
| Modality | text |
| Context | 1,000,000 tokens |
| Max output | — |
| Reasoning | Yes |
| Tool calling | Yes |
| Structured output | Not recorded |
| Open weights | Yes |
| Licence | NVIDIA Open Model License |
No verified Cybatar-run benchmark result has been published for this model yet.
Official-source release record for the exact Nemotron 3.5 Lightning model version.
Evidence ↗NVIDIA expands Nemotron with a 30B/3B-active open MoE model optimized for long-running agentic workloads.
Evidence ↗