Run Gemma 3 27B on M3 Max

Yes — Gemma 3 27B (27B) runs at 28 tok/s on M3 Max with 96 GB RAM using Q4_K_M quantization via MLX. First token latency is 1.3s. Google's 27B dense model, best-in-class reasoning for 24–48 GB Macs.

Speed

tok/s

First Token

1.3

seconds

RAM Needed

GB minimum

Engine

MLX

recommended

Benchmark Details

LLMCheck measured Gemma 3 27B on M3 Max using the standard methodology: Q4_K_M quantization, 256-token input, 512-token output, 3 runs averaged on a freshly-booted system.

Metric	Value
Tokens per second	28 tok/s
Time to first token	1.3s
Quantization	Q4_K_M
Minimum RAM	96 GB
Recommended engine	MLX
Parameters	27B
Benchmark date	2026-01

Q4_K_M 27B MLX M3 Max

Setup Guide: Run Gemma 3 27B on M3 Max

The recommended engine for Gemma 3 27B on M3 Max is MLX. Install with pip and pull the model:

pip install mlx-lm

mlx_lm.generate --model mlx-community/gemma-3-27b-q4_k_m --prompt "Hello!"

Alternatively, you can use Ollama for a simpler setup:

ollama run gemma3:27b

Performance on Other Apple Silicon Chips

Chip	Speed	First Token	Min RAM	Engine
M5 Max	42 tok/s	0.9s	64 GB	Ollama
M4 Max	35 tok/s	1.1s	48 GB	MLX
M4 Pro	25 tok/s	1.5s	24 GB	LM Studio

System Requirements

To run Gemma 3 27B on M3 Max you need:

• Mac with M3 Max chip (or newer)
• 96 GB unified memory minimum
• macOS 13 Ventura or later
• ~81-96 GB free disk space for the model file
• MLX installed (see our Ollama install guide)

Compare More Models

See how Gemma 3 27B stacks up against other models on your specific Mac hardware.

Open Compare Tool Full Leaderboard