qwen3.5-4b-mlx
LM-STUDIO MLXqwen3_5 · 4B · 4bit
Mar 6, 2026 · Apple M4 Pro
qwen3:14b
OLLAMA GGUFqwen3 · 14.8B · Q4_K_M
Jun 13, 2026 · AMD Ryzen 9 3900X 12-Core Processor
Global Score
62 vs 82
Hardware Fit
35 vs 90
Quality Score
73 vs 79
Hardware
qwen3.5-4b-mlx qwen3:14b
MachineMac miniMicro-Star International Co., Ltd. MS-7C37
CPUApple M4 ProAMD Ryzen 9 3900X 12-Core Processor
Cores1424
RAM64 GB64 GB
GPUApple M4 ProGameViewer Virtual Display Adapter, NVIDIA GeForce RTX 3080
OSmacOS 15.7.4Microsoft Windows 11 专业版 10.0.26200
Archarm64x64
Power Modebalancedbalanced
Performance
qwen3.5-4b-mlx qwen3:14b
Tokens/sec0.074.4
First chunk5548 ms324 ms
TTFT5.5 s2.6 s
Load timeN/A5.1 s
Memory usage2.9 GB9.0 GB
Memory %5%14%
HW Fit Score Breakdown
qwen3.5-4b-mlx
Speed
0/50
TTFT
6/20
Memory
29/30
qwen3:14b
Speed
50/50
TTFT
10/20
Memory
30/30
Quality
qwen3.5-4b-mlx
Reasoning
17/20
Coding
17/20
Instruction
9/20
Structured
8/15
Math
13/15
Multilingual
9/10
Reasoning: Strong Coding: Strong Instruction Following: Weak Structured Output: Adequate Math: Strong Multilingual: Strong
qwen3:14b
Reasoning
18/20
Coding
11/20
Instruction
17/20
Structured
14/15
Math
10/15
Multilingual
9/10
Reasoning: Strong Coding: Adequate Instruction Following: Strong Structured Output: Strong Math: Adequate Multilingual: Strong
Run yours and compare
$
npm install -g metrillm@latest$
metrillmRequires Node 20+ and Ollama or LM Studio running
Or run without installing: npx metrillm@latest