qwen3.5-9b-mlx

qwen3_5 · 9B · 4bit

Mac mini (Apple M4 Pro)

64 GB · macOS 15.7.4

Tested on March 12, 2026 · Submitted by alpha186@pm.me

Top 24% Compare

Global Score

77 /100

Good

Hardware Fit

67/100

Quality

81/100

Get this model

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

MLX versions & conversions

Hardware

Machine: Mac mini
CPU: Apple M4 Pro
Cores: 14 total (10 perf + 4 eff)
Frequency: 2.4 GHz
RAM: 64 GB LPDDR5
GPU: Apple M4 Pro
OS: macOS 15.7.4
Arch: arm64
Power Mode: balanced

Performance

Tokens/sec Tokens generated per second — higher is better: 20.3
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±25.0
First chunk latency Delay before the first streamed chunk arrives from runtime: 4.2 s
Time to first token How fast the model starts responding — lower is better: 4.2 s
Load time Time to load the model into memory before inference: N/A
Memory usage RAM consumed during inference — percentage of total system memory: 5.6 GB (9%)
Total tokens Total number of tokens generated across all benchmark prompts: 602

Score breakdown

Speed

29/50

Time to first token

8/20

Memory

30/30

Quality

Reasoning

15/20

Coding

18/20

Instruction following

16/20

Structured output

14/15

Math

8/15

Multilingual

10/10

Category levels

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

Metadata

Spec version: 0.2.1
Runtime: LM Studio 0.4.6+1
Model format: MLX
Hardware profile: HIGH-END
Result hash: 21faacef46b07a3f1041791103b550616e8a20c4f6a33d4753eb5ad8747116a7

Interpretation

Hardware fit: 67/100. Overall suitability: GOOD (Global 77/100). Category profile: Reasoning: Adequate, Coding: Strong, Instruction Following: Strong, Structured Output: Strong, Math: Adequate, Multilingual: Strong.

Warnings

Token speed is unstable (stddev 25.0 tok/s, mean 20.3 tok/s) — may indicate thermal throttling or memory pressure.

Bench Environment

Power: AC CPU load: avg 24% (peak 34%)

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