Rent a GPU

Can the Apple M3 Pro (36GB) run Gemma 4 31B?

Apple M3 Pro (36GB) × Gemma 4 31B8K context
YesRuns fully on the GPU
Memory at Q5_K_M24 GB of 24.1 GB
Weights 22.3 GBKV cache 1.2 GBRuntime 0.6 GB
QuantQ5_K_MVery good
Speed~4tokens per second
Max context9Ktokens that fit at Q5_K_M
  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 M3 Pro (36GB) can run Gemma 4 31B fully in unified memory at Q5_K_M (very good), using about 24 GB of the 24.1 GB available, at roughly 4 tokens/second.

Try other context lengths

Run it with Ollama

With Ollama installed, one command downloads and starts Gemma 4 31B:

ollama run gemma4:31b

Ollama's default download is Q4_K_M (about 20.8 GB here). You have room for Q5_K_M, which is very good: pick that GGUF in LM Studio or from Hugging Face.

New to this? Step-by-step setup for the Apple M3 Pro (36GB)

Run it in the cloud

No Apple M3 Pro (36GB)? Run Gemma 4 31B 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 24.1 GB the Apple M3 Pro (36GB) can use. macOS lets the GPU use about 24.1 GB of the 36 GB unified memory by default.

QuantQualityMemoryFits?Speed
FP16Full quality64.4 GBToo big—
Q8_0Near-lossless35 GBToo big—
Q6_KExcellent27.4 GBToo big—
Q5_K_MVery good24 GBFits~4 tok/s
Q4_K_MGood, most popular20.8 GBFits~5 tok/s
Q3_K_MNoticeably degraded17.1 GBFits~6 tok/s
Q2_KHeavily degraded13.5 GBFits~8 tok/s

Measured speed from real Apple M3 Pro (36GB) owners

Tokens per second that people measured on their own Apple M3 Pro (36GB), next to our estimate.

No one has measured Gemma 4 31B on the Apple M3 Pro (36GB) yet. The speeds above are estimates. Run the benchmark and yours becomes the first real number on this page.

Have an Apple M3 Pro (36GB)? Add your real speed

One command times Gemma 4 31B 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 -- gemma4:31b

Windows (PowerShell)

$env:GPUFIT_MODEL="gemma4:31b"; irm https://gpufit.dev/bench.ps1 | iex

How the benchmark works

GPUs that run Gemma 4 31B comfortably

Models that run well on the Apple M3 Pro (36GB)

Frequently asked questions

Can Apple M3 Pro (36GB) run Gemma 4 31B?

Yes. The Apple M3 Pro (36GB) can run Gemma 4 31B fully in unified memory at Q5_K_M (very good), using about 24 GB of the 24.1 GB available, at roughly 4 tokens/second.

How much VRAM does Gemma 4 31B need?

About 20.8 GB at Q4_K_M with 8,192 tokens of context, and about 64.4 GB at full FP16 precision.

How fast is Gemma 4 31B on the Apple M3 Pro (36GB)?

Roughly 4 tokens per second at Q5_K_M, estimated from the Apple M3 Pro (36GB)'s 150 GB/s memory bandwidth.