Run Qwen 3 4B on M6

Qwen 3 4B runs at 53 tok/s (53.35 as recorded) on a 32 GB M6, a published figure reported by tbreak.com (MLX, quantization not stated). It needs about 3 GB of unified memory. It is Alibaba's 4B dense model, released under the Apache 2.0 licence.

Speed

sourced53

tok/s

First Token

—

seconds

RAM Needed

GB minimum

Engine

MLX

of this figure

Want to run Qwen 3 4B faster — or step up to bigger models? Find your Mac →

The figure

This is a published figure: a real run, reported by its source and linked below. The index records it as given and never adjusts it.

Metric	Value
Tokens per second	sourced53 tok/s
Provenance	published figure · tbreak.com
Quantization	quantization not stated
Engine	MLX
Figure dated	2026-09-21
Minimum RAM	3 GB
Parameters	4B
Chip bandwidth	170 GB/s

Every figure for Qwen 3 4B on the M6

Each row as the index holds it, with its source. “Mac RAM” is the memory of the Mac a run used, not the model’s requirement (3 GB). The headline figure follows the reference rule: the lower-median measured plain-decode run at 5 bits or fewer (a run whose quant was not stated counts only when no stated-quant run qualifies), else the estimate; a run marked “not used to anchor estimates” is listed here but never becomes the headline.

Speed	Quant	Engine	Mac RAM	Context	Source	Note
sourced53.35 tok/s	quantization not stated	MLX	32 GB	—	tbreak.com	Mac mini M6, 12-core GPU. tbreak review: MLX 0.32.2 / mlx-lm 0.31.3; quantization not stated.
community31.47 tok/s	8-bit	LM Studio	24 GB	short prompt	nyaa.sh	Mac mini M6. Qwen3-4B-Thinking-2507; LM Studio (engine not stated); single run.

Provenance key

sourced

Published figure — vendor or press review

estimated

Derived from published bandwidth model

community

Linked individual run — blog, forum or submission

Setup: run Qwen 3 4B on a Mac

Ollama — install it (brew install ollama or the app from ollama.com), then pull the model by its tag, checked on ollama.com on 2026-10-17:

ollama run qwen3:4b

MLX — pick a build from Hugging Face’s MLX listings, then run it with its repo id:

pip install -U mlx-lm

mlx_lm.generate --model <repo-id> --prompt "Hello"

Downloads: Search the Ollama library → · model card →

The 53 tok/s figure above came from MLX at quantization not stated; other runtimes and quants run at different speeds.

Prefer a GUI? LM Studio downloads GGUF and MLX builds in one click. Step by step: install Ollama on a Mac.

Qwen 3 4B on other Apple Silicon chips

Chip	Speed	Quant	Engine	Bandwidth
M5 Max	estimated161 tok/s	Q4_K_M	Ollama	614 GB/s
M4 Pro	estimated82 tok/s	Q4_K_M	MLX	273 GB/s
M2	estimated33 tok/s	Q4_K_M	Ollama	100 GB/s

About Qwen 3 4B

Alibaba's 4B model, ~161 tok/s on M5 Max. Apache 2.0 license, LLMCheck capability 11/50.

Apache 2.0 licence · 4B · 128K context · LLMCheck capability 11/50 · see its leaderboard rank →

System requirements

• 3 GB of unified memory or more (the index’s minimum: the weights at the quant that fits, plus ~2 GB)
• About 1 GB of free disk space for the model file
• Any Apple Silicon Mac with that much memory runs it; the M6’s 170 GB/s memory bandwidth sets the speed shown here
• A current runtime: Ollama, LM Studio, llama.cpp or MLX (Ollama install guide)

Macs with the M6

Mac mini M6 (16 GB) fits · 153 GB/s
Mac mini M6 (24 GB) fits · 170 GB/s
Mac mini M6 (32 GB) fits · 170 GB/s
Best local LLMs for a 8 GB Mac the smallest RAM tier that fits Qwen 3 4B

🛒 Get a Mac that runs Qwen 3 4B

Qwen 3 4B needs about 3 GB of unified memory. These current Apple Silicon Macs have the headroom to run it comfortably:

Mac mini M6 on Amazon → MacBook Air M4 on Amazon →

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

Compare More Models

See how Qwen 3 4B stacks up against other models on your specific Mac hardware.

Open Compare Tool Full Leaderboard