Run Devstral Small 24B on M5 Max

Yes — Devstral Small 24B (24B) runs at 45 tok/s on M5 Max with 128 GB RAM using Q4_K_M quantization via MLX. First token latency is 0.6s. A capable open-source LLM with 24B parameters.

Speed

tok/s

First Token

0.6

seconds

RAM Needed

128

GB minimum

Engine

MLX

recommended

Benchmark Details

LLMCheck measured Devstral Small 24B on M5 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	45 tok/s
Time to first token	0.6s
Quantization	Q4_K_M
Minimum RAM	128 GB
Recommended engine	MLX
Parameters	24B
Benchmark date	2026-05

Q4_K_M 24B MLX M5 Max

Setup Guide: Run Devstral Small 24B on M5 Max

The recommended engine for Devstral Small 24B on M5 Max is MLX. Install with pip and pull the model:

pip install mlx-lm

mlx_lm.generate --model mlx-community/devstral-small-24b-q4_k_m --prompt "Hello!"

Alternatively, you can use Ollama for a simpler setup:

ollama run devstral-small-24b

Performance on Other Apple Silicon Chips

Chip	Speed	First Token	Min RAM	Engine
M4 Max	35 tok/s	0.8s	48 GB	MLX
M4 Pro	26 tok/s	1.0s	32 GB	Ollama

System Requirements

To run Devstral Small 24B on M5 Max you need:

• Mac with M5 Max chip (or newer)
• 128 GB unified memory minimum
• macOS 13 Ventura or later
• ~108-128 GB free disk space for the model file
• MLX installed (see our Ollama install guide)

Compare More Models

See how Devstral Small 24B stacks up against other models on your specific Mac hardware.

Open Compare Tool Full Leaderboard