Run Qwen 3 4B on M2

Yes — Qwen 3 4B (4B) runs at 52 tok/s on M2 with 8 GB RAM using Q4_K_M quantization via Ollama. First token latency is 0.7s. Alibaba's 4B dense model with strong multilingual and coding capabilities.

Speed
52
tok/s
First Token
0.7
seconds
RAM Needed
8
GB minimum
Engine
Ollama
recommended

Benchmark Details

LLMCheck measured Qwen 3 4B on M2 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 second52 tok/s
Time to first token0.7s
QuantizationQ4_K_M
Minimum RAM8 GB
Recommended engineOllama
Parameters4B
Benchmark date2026-01

Q4_K_M 4B Ollama M2

Setup Guide: Run Qwen 3 4B on M2

The recommended engine for Qwen 3 4B on M2 is Ollama. Install Ollama, then pull the model:

ollama run qwen3:4b

Ollama handles quantization automatically — it will download the Q4_K_M variant (~8 GB) and start an interactive chat session.

Performance on Other Apple Silicon Chips

ChipSpeedFirst TokenMin RAMEngine
M5 Max 135 tok/s 0.2s 64 GB Ollama
M4 Pro 118 tok/s 0.3s 24 GB MLX

System Requirements

To run Qwen 3 4B on M2 you need:

Compare More Models

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

Open Compare Tool Full Leaderboard