Written by Jakub Rusinowski · Last updated September 19, 2026
NVIDIA's agentic line, built on a hybrid stack that interleaves Mamba-2 state-space layers and MoE blocks with a minority of attention layers. Only the attention layers hold a KV cache, which is what lets it advertise a 1M-token context at this size.
| Licence | What it permits | Applies to |
|---|---|---|
OpenMDW-1.1 | Commercial use permitted Commercial use permitted. No usage restrictions beyond attribution. | Nemotron 3.5 Lightning 30B-A3B |
| Nemotron 3.5 Lightning 30B-A3B | Min 19 GB VRAM · Q4_K_M · 1,000,000 ctx · ollama run nemotron-3.5-lightning:30b |
The cheapest GPU that runs Nemotron 3.5 locally (min 19 GB VRAM) is the AMD Radeon RX 7900 XT (20 GB).
Install Ollama then run: ollama run nemotron-3.5-lightning:30b
Minimum VRAM: 19 GB. For best results use Q4_K_M quantization.
Nemotron 3.5 needs about 19 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: Nemotron 3.5 Lightning 30B-A3B (19 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Yes — Nemotron 3.5 runs on an RTX 4090 (24 GB) and other 24 GB cards such as the RTX 3090. Smaller variants also fit comfortably on 8–16 GB GPUs at Q4_K_M.
Q4_K_M is the best balance of quality and VRAM for Nemotron 3.5 in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
Install Ollama, then run: ollama run nemotron-3.5-lightning:30b. This downloads Nemotron 3.5 and starts a local, OpenAI-compatible endpoint — no internet connection is needed after the initial download.