Run Mistral Medium 4 on M4 Max

Yes — Mistral Medium 4 (41B) runs at 38 tok/s on M4 Max with 48 GB RAM using Q4_K_M quantization via MLX. First token latency is 0.8s. A capable open-source LLM with 41B parameters.

Speed
38
tok/s
First Token
0.8
seconds
RAM Needed
48
GB minimum
Engine
MLX
recommended

Benchmark Details

LLMCheck measured Mistral Medium 4 on M4 Max using the standard methodology: Q4_K_M quantization, 256-token input, 512-token output, 3 runs averaged on a freshly-booted system.

MetricValue
Tokens per second38 tok/s
Time to first token0.8s
QuantizationQ4_K_M
Minimum RAM48 GB
Recommended engineMLX
Parameters41B
Benchmark date2026-07

Q4_K_M 41B MLX M4 Max

Setup Guide: Run Mistral Medium 4 on M4 Max

The recommended engine for Mistral Medium 4 on M4 Max is MLX. Install with pip and pull the model:

pip install mlx-lm
mlx_lm.generate --model mlx-community/mistral-medium-4-q4_k_m --prompt "Hello!"

Alternatively, you can use Ollama for a simpler setup:

ollama run mistral-medium-4

Performance on Other Apple Silicon Chips

ChipSpeedFirst TokenMin RAMEngine
M5 Max 48 tok/s 0.6s 128 GB MLX
M5 Max 42 tok/s 0.7s 64 GB Ollama
M3 Max 30 tok/s 1.0s 64 GB Ollama

System Requirements

To run Mistral Medium 4 on M4 Max you need:

🛒 Get a Mac that runs Mistral Medium 4

Mistral Medium 4 needs about 48 GB of unified memory. These current Apple Silicon Macs have the headroom to run it comfortably:

Not sure which Mac fits your budget? See the best Mac for running this →

As an Amazon Associate, LLMCheck earns from qualifying purchases. These affiliate links cost you nothing extra and help keep our benchmarks free.

Compare More Models

See how Mistral Medium 4 stacks up against other models on your specific Mac hardware.

Open Compare Tool Full Leaderboard