qwen3.5-27b
LM-STUDIO MLXqwen3_5 · 27B · 4bit
Mar 3, 2026 · Apple M4
mlx-community/gemma-3-4b-it-qat-4bit
LM-STUDIO GGUFMar 5, 2026 · Apple M4
Global Score
75 vs 80
Hardware Fit
54 vs 100
Quality Score
84 vs 71
Hardware
qwen3.5-27b mlx-community/gemm…
MachineMacBook AirMacBook Air
CPUApple M4Apple M4
Cores1010
RAM32 GB32 GB
GPUApple M4Apple M4
OSmacOS 26.3macOS 26.3
Archarm64arm64
Power Modebalancedbalanced
Performance
qwen3.5-27b mlx-community/gemm…
Tokens/sec6.643.2
First chunk4364 ms428 ms
TTFT4.4 s428 ms
Load timeN/AN/A
Memory usage15.0 GB1.3 GB
Memory %47%4%
HW Fit Score Breakdown
qwen3.5-27b
Speed
13/50
TTFT
11/20
Memory
30/30
mlx-community/gemma-3-4…
Speed
50/50
TTFT
20/20
Memory
30/30
Quality
qwen3.5-27b
Reasoning
17/20
Coding
18/20
Instruction
11/20
Structured
15/15
Math
13/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Strong Multilingual: Strong
mlx-community/gemma-3-4…
Reasoning
9/20
Coding
17/20
Instruction
17/20
Structured
13/15
Math
5/15
Multilingual
10/10
Reasoning: Weak Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Weak 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