Rent a GPU

Can the Apple M4 Max (64GB) run Nemotron 3 Nano 30B-A3B?

Apple M4 Max (64GB) × Nemotron 3 Nano 30B-A3B8K context
YesRuns fully on the GPU
Memory at Q8_034.2 GB of 48 GB
Weights 33.6 GBKV cache 0.1 GBRuntime 0.6 GB
QuantQ8_0Near-lossless
Speed~88tokens per second
Max context256Kthe model's full window
  1. FP16
  2. Q8_0
  3. Q6_K
  4. Q5_K_M
  5. Q4_K_M
  6. Q3_K_M
  7. Q2_K

Yes. The Apple M4 Max (64GB) can run Nemotron 3 Nano 30B-A3B fully in unified memory at Q8_0 (near-lossless), using about 34.2 GB of the 48 GB available, at roughly 88 tokens/second.

Try other context lengths

Run it with Ollama

With Ollama installed, one command downloads and starts Nemotron 3 Nano 30B-A3B:

ollama run nemotron-3-nano:30b

Ollama's default download is Q4_K_M (about 19.8 GB here). You have room for Q8_0, which is near-lossless: pick that GGUF in LM Studio or from Hugging Face.

New to this? Step-by-step setup for the Apple M4 Max (64GB)

Run it in the cloud

No Apple M4 Max (64GB)? Run Nemotron 3 Nano 30B-A3B in the cloud.

Rent a cloud GPU by the hour. No upfront cost, stop any time.

Rent a GPU on RunPodRent a GPU on Vast.ai

Memory needed by quantization

At 8,192 tokens of context. The black line marks the 48 GB the Apple M4 Max (64GB) can use. macOS lets the GPU use about 48 GB of the 64 GB unified memory by default.

QuantQualityMemoryFits?Speed
FP16Full quality63.9 GBToo big—
Q8_0Near-lossless34.2 GBFits~88 tok/s
Q6_KExcellent26.6 GBFits~114 tok/s
Q5_K_MVery good23.1 GBFits~132 tok/s
Q4_K_MGood, most popular19.8 GBFits~154 tok/s
Q3_K_MNoticeably degraded16.1 GBFits~192 tok/s
Q2_KHeavily degraded12.5 GBFits~250 tok/s

Measured speed from real Apple M4 Max (64GB) owners

Tokens per second that people measured on their own Apple M4 Max (64GB), next to our estimate.

No one has measured Nemotron 3 Nano 30B-A3B on the Apple M4 Max (64GB) yet. The speeds above are estimates. Run the benchmark and yours becomes the first real number on this page.

Have an Apple M4 Max (64GB)? Add your real speed

One command times Nemotron 3 Nano 30B-A3B in your own Ollama, shows you exactly what it would send, and asks before sending anything. No sign-up, no personal data.

macOS and Linux

curl -fsSL https://gpufit.dev/bench.sh | sh -s -- nemotron-3-nano:30b

Windows (PowerShell)

$env:GPUFIT_MODEL="nemotron-3-nano:30b"; irm https://gpufit.dev/bench.ps1 | iex

How the benchmark works

GPUs that run Nemotron 3 Nano 30B-A3B comfortably

Models that run well on the Apple M4 Max (64GB)

Frequently asked questions

Can Apple M4 Max (64GB) run Nemotron 3 Nano 30B-A3B?

Yes. The Apple M4 Max (64GB) can run Nemotron 3 Nano 30B-A3B fully in unified memory at Q8_0 (near-lossless), using about 34.2 GB of the 48 GB available, at roughly 88 tokens/second.

How much VRAM does Nemotron 3 Nano 30B-A3B need?

About 19.8 GB at Q4_K_M with 8,192 tokens of context, and about 63.9 GB at full FP16 precision.

How fast is Nemotron 3 Nano 30B-A3B on the Apple M4 Max (64GB)?

Roughly 88 tokens per second at Q8_0, estimated from the Apple M4 Max (64GB)'s 546 GB/s memory bandwidth.