OPEN WEIGHTSOFFICIAL GGUF30B TOTAL · 3B ACTIVEVERIFIED AUG 13, 2026

Nemotron 3.5 Lightning local requirements

Short answer: the official Q4_0 GGUF is 18.9GB and needs about 20.5GB at 4K context before the operating system's reserve. A 24GB GPU is a tight full-offload tier; 32GB GPU or unified memory is the comfortable local target.

30B / 3B
total / active parameters
18.9GB
official Q4_0 file
~20.5GB
estimated Q4 working set @4K
1M
maximum supported context

Answer by hardware

These are memory-fit estimates, not measured speed claims. Longer context and GPU/CPU splitting need more headroom.

Phone / 24GB system
No
The official Q4 file alone is larger than the available memory budget.
24GB GPU
Runs, but tight
The current Q4_0 fits at short context; close other GPU-heavy apps.
32GB GPU
Recommended Q4 tier
Useful allocation headroom for agents and longer prompts.
32GB Mac
Recommended
Q4 fits with practical macOS and runtime headroom.
48GB Mac
Comfortable
Room for Q4 at long context or the larger Q8_0 quant.
64GB+
Q8 or long context
The better tier for Q8_0, heavy agents, and large-context experiments.

Official GGUF sizes and working memory

Download sizes come from the official ggml-org repository. Working set adds 5% loaded-weight overhead, the published hybrid-attention KV layout at 4K context, and a small runtime allowance.

QuantDownloadEst. working set @4KPractical tier
Q4_018.9 GB~20.5 GB24GB VRAM / 32GB Mac preferred
Q8_033.6 GB~35.9 GB48GB+ memory

What the 1M context claim costs locally

Nemotron interleaves Mamba-2, MoE, and six full-attention layers. That keeps its KV cache far smaller than a conventional 30B transformer, but processing a million-token prompt still needs substantial temporary memory and time.

ContextAttention KV cacheQ4 working setInterpretation
4K~0 GB~20.5 GBConsumer-local range
32K~0.2 GB~20.6 GBConsumer-local range
256K~1.5 GB~21.9 GBHigh-memory experiment
1M~6 GB~26.4 GBDatacenter-class target

Why “3B active” does not mean “3B memory”

Mixture-of-Experts routing selects roughly 3B parameters for each token, which can make decoding much faster than a dense 30B model. It does not discard the other experts: the complete 30B model still has to be stored and addressable. Active parameters describe compute per token; the GGUF file describes the weight memory.

See every phone and Mac result →

Primary sources

NVIDIA model card — architecture, 30B/3B parameters, 1M context, license, and validated serving configurations ↗Official ggml-org GGUF — file-derived Q4_0 and Q8_0 sizes ↗

FAQ

How much RAM does Nemotron 3.5 Lightning need?

The official Q4_0 GGUF is 18.9GB. Our short-context working-set estimate is about 20.5GB before operating-system reserve, so 24GB is the minimum sensible memory tier and 32GB is the comfortable recommendation.

Can Nemotron 3.5 Lightning run on a 24GB GPU?

Yes at the current Q4_0 quant, whose weights are 18.9GB. The estimated 4K working set is about 20.5GB before OS reserve, so close other GPU workloads and begin with a short context.

Can a 32GB Mac run Nemotron 3.5 Lightning?

Yes. The current Q4_0 quant fits at short context with useful headroom. A 24GB Mac is tighter after macOS reserve; 32GB is the safer target.

Does 3B active mean it uses only 3B-model memory?

No. The model routes each token through roughly 3B active parameters, which reduces compute, but all 30B parameters still have to live in RAM, VRAM, or a split CPU/GPU working set.

Can a phone run Nemotron 3.5 Lightning?

No current phone has enough app-usable memory for the official 18.9GB Q4_0 file plus runtime memory. The 3B-active label is not a phone-memory requirement.

Can local hardware really use the 1M-token context?

The model supports up to 1M tokens, but that is not the practical default for a consumer device. The attention KV cache alone grows to about 6GB at 1M, before prompt processing buffers and the 18.9GB Q4 weights; NVIDIA validates 1M on B200-class hardware.