qwen3.5-4b-mlx
LM-STUDIO MLXqwen3_5 · 4B · 4bit
Mar 6, 2026 · Apple M4 Pro
gemma3:270m
OLLAMA GGUFgemma3 · 268.10M · Q8_0
Mar 5, 2026 · Apple M3 Pro
Global Score
62 vs 50
Hardware Fit
35 vs 97
Quality Score
73 vs 30
Hardware
qwen3.5-4b-mlx gemma3:270m
MachineMac miniMacBook Pro
CPUApple M4 ProApple M3 Pro
Cores1412
RAM64 GB36 GB
GPUApple M4 ProApple M3 Pro
OSmacOS 15.7.4macOS 15.7.5
Archarm64arm64
Power Modebalancedbalanced
Performance
qwen3.5-4b-mlx gemma3:270m
Tokens/sec0.0239.2
First chunk5548 ms127 ms
TTFT5.5 s127 ms
Load timeN/A0.7 s
Memory usage2.9 GB0.6 GB
Memory %5%2%
HW Fit Score Breakdown
qwen3.5-4b-mlx
Speed
0/50
TTFT
6/20
Memory
29/30
gemma3:270m
Speed
50/50
TTFT
20/20
Memory
27/30
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
gemma3:270m
Reasoning
4/20
Coding
2/20
Instruction
10/20
Structured
8/15
Math
1/15
Multilingual
5/10
Reasoning: Poor Coding: Poor Instruction Following: Adequate Structured Output: Adequate Math: Poor Multilingual: Weak
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