qwen3:4b

qwen3 · 4.0B · Q4_K_M

Apple M4

32 GB · macOS 26.3

Tested on March 1, 2026

Top 89% Compare

Global Score

48 /100

Marginal

Hardware Fit

53/100

Quality

45/100

Get this model

🦙

Ollama

ollama pull qwen3:4b

View on Ollama Library

ollama.com/library/qwen3

🤗

Find on HuggingFace

GGUF versions & quantizations

Hardware

CPU: Apple M4
Cores: 10 total (4 perf + 6 eff)
Frequency: 2.4 GHz
RAM: 32 GB LPDDR5
GPU: Apple M4
OS: macOS 26.3
Arch: arm64

Performance

Tokens/sec Tokens generated per second — higher is better: 11.3
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±0.4
Time to first token How fast the model starts responding — lower is better: 9.2 s
Load time Time to load the model into memory before inference: 0.1 s
Memory usage RAM consumed during inference — percentage of total system memory: 7.2 GB (23%)
Total tokens Total number of tokens generated across all benchmark prompts: 1310

Score breakdown

Speed

17/40

Time to first token

6/30

Memory

30/30

Quality

Reasoning

4/20

Coding

6/20

Instruction following

8/20

Structured output

6/15

Math

12/15

Multilingual

9/10

Category levels

Reasoning: Poor Coding: Weak Instruction Following: Weak Structured Output: Weak Math: Strong Multilingual: Strong

Metadata

Spec version: 0.1.0
Runtime: Ollama 0.17.4
Model format: GGUF
Hardware profile: BALANCED
Result hash: b6a0712716e79be0c2262bd025f1bfb30599a7578bc2b6d1e06eb66d3454aadd

Interpretation

Hardware fit: 53/100. Overall suitability: MARGINAL (Global 48/100). Category profile: Reasoning: Poor, Coding: Weak, Instruction Following: Weak, Structured Output: Weak, Math: Strong, Multilingual: Strong.

Run yours now

npx metrillm@latest bench

Requires Node 20+ and Ollama running