Can I Run Gemma 4 on Mac Studio (M3 Ultra, 256 GB)?

Written by Jakub Rusinowski · Last updated June 3, 2026

Mac Studio (M3 Ultra, 256 GB) — what it gives a model

Usable memory for models	192 GB
Memory bandwidth	819 GB/s
Form factor	Desktop
Operating system	macOS
Memory upgradeable	No — soldered

Gemma 4 on Mac Studio (M3 Ultra, 256 GB): memory by quantization

Quant	Memory needed	Fits 192 GB?	Max context	Est. speed	Download
F16	64.6 GB	✓ Yes	128K	~7 tok/s	62 GB
Q8_0	35.5 GB	✓ Yes	128K	~12.8 tok/s	32.9 GB
Q6_K	28 GB	✓ Yes	128K	~16.2 tok/s	25.4 GB
Q5_K_M	24.6 GB	✓ Yes	128K	~18.5 tok/s	22 GB
Q4_K_M	21.3 GB	✓ Yes	128K	~21.4 tok/s	18.7 GB
Q3_K_M	15.8 GB	✓ Yes	128K	~29 tok/s	13.2 GB
Q2_K	12.8 GB	✓ Yes	128K	~36.1 tok/s	10.2 GB

Which Gemma 4 sizes fit

Variant	Needs at Q4_K_M	Fits?	Est. speed
Gemma 4 31B	21.3 GB	✓ Fits	~21.4 tok/s
Gemma 4 26B-A4B	18.2 GB	✓ Fits	~97.2 tok/s
Gemma 4 12B (Unified)	9.4 GB	✓ Fits	~48.6 tok/s
Gemma 4 E4B	6.8 GB	✓ Fits	~66.5 tok/s
Gemma 4 E2B	4.9 GB	✓ Fits	~91.1 tok/s

What to watch out for

• This model does not publish its full attention configuration, so the KV-cache share of these figures is inferred from its parameter count rather than computed exactly.
• Memory on this machine is not upgradeable, so the configuration you buy is the ceiling for every model you will ever run on it.

Mac Studio M3 Ultra 256 GB limitations

• The largest single-box unified memory pool available, but at roughly half the memory bandwidth of a high-end discrete GPU — very large models load, then run slowly.
• Memory is soldered; the configuration chosen at purchase is permanent.

Recommended setup

Ollama or LM Studio (Metal) — MLX for the fastest Apple-native throughput

How these numbers are calculated

• Memory need = quantized weights + KV cache at 8,192 tokens (f16) + 0.8 GB runtime overhead.
• 256 GB unified memory at 819 GB/s, shared between CPU and GPU.
• macOS reserves a share of unified memory for the system, so not all of it is available to a model.
• Throughput is a memory-bandwidth roofline estimate, not a measurement. It is labelled "estimated" everywhere it appears.
• KV cache is inferred from the parameter count because this model does not publish its full attention configuration.

FAQ

Can I run Gemma 4 on Mac Studio (M3 Ultra, 256 GB)?

Yes, comfortably — Gemma 4 31B at Q8_0 needs about 35.5 GB of the 192 GB usable on Mac Studio (M3 Ultra, 256 GB), leaving ~156.5 GB spare and running at ~12.8 tok/s (estimated), with room for about 131,072 tokens of context.

Which quantization of Gemma 4 should I use on Mac Studio (M3 Ultra, 256 GB)?

Q8_0 — it needs about 35.5 GB of the 192 GB available, downloads as roughly 32.9 GB, and runs at an estimated 12.8 tokens/sec with up to 128K of context.

What limits Gemma 4 on Mac Studio (M3 Ultra, 256 GB)?

Nothing binding — the model fits with headroom and generates at a usable speed on this hardware.

Which runtime should I use?

Ollama or LM Studio (Metal) — MLX for the fastest Apple-native throughput

Other Computers

• Gemma 4 on RTX 4090 desktop
• Gemma 4 on RTX 5090 desktop
• Gemma 4 on RTX 3090 desktop
• Gemma 4 on Mac Studio M4 Max 64 GB

Other Models on Mac Studio (M3 Ultra, 256 GB)

• GLM-4.7 / GLM-Z1 on Mac Studio (M3 Ultra, 256 GB)
• GLM-5 / GLM-5.1 on Mac Studio (M3 Ultra, 256 GB)
• GLM-6 on Mac Studio (M3 Ultra, 256 GB)
• GPT-OSS on Mac Studio (M3 Ultra, 256 GB)
• Granite 3.0 on Mac Studio (M3 Ultra, 256 GB)

Gemma 4 on GPUs

• Gemma 4 on NVIDIA GeForce RTX 5090
• Gemma 4 on NVIDIA GeForce RTX 5080
• Gemma 4 on NVIDIA GeForce RTX 5070 Ti
• Gemma 4 on NVIDIA GeForce RTX 5070

What This Model Is Good At

• Best local LLMs for coding
• Best local LLMs for reasoning
• Best local LLMs for rag

Model & Tools

• Gemma 4 model page
• VRAM calculator
• Check your own hardware

← Can I Run It? | Gemma 4 model page | Check your hardware