qwen3.5:4b

qwen35 · 4.7B · Q4_K_M

THINKING MODEL

Mac mini (Apple M4)

16 GB · macOS 26.3

Tested on March 5, 2026 · Submitted by susutechno

Top 83% Compare

Global Score

48 /100

Marginal

Hardware Fit

77/100

Quality

35/100

Get this model

🦙

Ollama

ollama pull qwen3.5:4b

View on Ollama Library

ollama.com/library/qwen3.5

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

GGUF versions & quantizations

Hardware

Machine: Mac mini
CPU: Apple M4
Cores: 10 total (4 perf + 6 eff)
Frequency: 2.4 GHz
RAM: 16 GB LPDDR5
GPU: Apple M4
OS: macOS 26.3
Arch: arm64
Power Mode: balanced

Performance

Tokens/sec Tokens generated per second — higher is better: 18.9
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±0.1
First chunk latency Delay before the first streamed chunk arrives from runtime: 316 ms
Time to first token How fast the model starts responding — lower is better: 13.0 s
Load time Time to load the model into memory before inference: 3.7 s
Memory usage RAM consumed during inference — percentage of total system memory: 5.9 GB (37%)
Total tokens Total number of tokens generated across all benchmark prompts: 1429
Thinking tokens (est.) Estimated internal reasoning tokens — models with "thinking" generate more tokens: ~768

Score breakdown

Speed

43/50

Time to first token

4/20

Memory

30/30

Quality

Reasoning

10/20

Coding

1/20

Instruction following

4/20

Structured output

3/15

Math

8/15

Multilingual

9/10

Category levels

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

Metadata

Spec version: 0.2.1
Runtime: Ollama 0.17.6
Model format: GGUF
Hardware profile: ENTRY
Result hash: cbb865ffbbbc6fa08ab929dc6f1b226decda2c8049c7ac527509c8a3472b5278

Interpretation

Hardware fit: 77/100. Overall suitability: MARGINAL (Global 48/100). Category profile: Reasoning: Weak, Coding: Poor, Instruction Following: Poor, Structured Output: Poor, Math: Adequate, Multilingual: Strong.

Bench Environment

Power: AC CPU load: avg 14% (peak 18%)

Run yours now

$ npm install -g metrillm@latest

$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest