qwen3.8-27b-mlx
LM-STUDIO MLXqwen3_5 · 27B · 8bit
Aug 17, 2026 · Apple M5 Max
qwen3.5:4b
OLLAMA GGUFqwen35 · 4.7B · Q4_K_M
Aug 25, 2026 · Apple M1 Pro
Global Score
59 vs 83
Hardware Fit
63 vs 99
Quality Score
58 vs 76
Hardware
qwen3.8-27b-mlx qwen3.5:4b
MachineMacBook ProMacBook Pro
CPUApple M5 MaxApple M1 Pro
Cores1810
RAM64 GB16 GB
GPUApple M5 MaxApple M1 Pro
OSmacOS 27.0macOS 26.6.2
Archarm64arm64
Power Modebalancedbalanced
Performance
qwen3.8-27b-mlx qwen3.5:4b
Tokens/sec18.336.6
First chunk4 ms204 ms
TTFT530 ms204 ms
Load timeN/A2.7 s
Memory usage38.5 GB5.3 GB
Memory %60%33%
HW Fit Score Breakdown
qwen3.8-27b-mlx
Speed
23/50
TTFT
20/20
Memory
20/30
qwen3.5:4b
Speed
50/50
TTFT
20/20
Memory
29/30
Quality
qwen3.8-27b-mlx
Reasoning
19/20
Coding
14/20
Instruction
6/20
Structured
1/15
Math
13/15
Multilingual
5/10
Reasoning: Strong Coding: Adequate Instruction Following: Weak Structured Output: Poor Math: Strong Multilingual: Weak
qwen3.5:4b
Reasoning
16/20
Coding
14/20
Instruction
15/20
Structured
15/15
Math
7/15
Multilingual
9/10
Reasoning: Strong Coding: Adequate 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