qwen3.8-27b-mlx
LM-STUDIO MLXqwen3_5 · 27B · 8bit
Aug 17, 2026 · Apple M5 Max
qwen2.5:14b
OLLAMA GGUFqwen2 · 14.8B · Q4_K_M
Sep 15, 2026 · Intel Core™ i5-14400F
Global Score
59 vs 90
Hardware Fit
63 vs 100
Quality Score
58 vs 85
Hardware
qwen3.8-27b-mlx qwen2.5:14b
MachineMacBook ProGigabyte Technology Co., Ltd. B760 AORUS ELITE AX DDR4
CPUApple M5 MaxIntel Core™ i5-14400F
Cores1816
RAM64 GB32 GB
GPUApple M5 MaxNVIDIA GeForce RTX 5060 Ti
OSmacOS 27.0Microsoft Windows 11 Professionnel 10.0.26200
Archarm64x64
Power Modebalancedbalanced
Performance
qwen3.8-27b-mlx qwen2.5:14b
Tokens/sec18.337.8
First chunk4 ms74 ms
TTFT530 ms74 ms
Load timeN/A6.2 s
Memory usage38.5 GB8.8 GB
Memory %60%28%
HW Fit Score Breakdown
qwen3.8-27b-mlx
Speed
23/50
TTFT
20/20
Memory
20/30
qwen2.5:14b
Speed
50/50
TTFT
20/20
Memory
30/30
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
qwen2.5:14b
Reasoning
17/20
Coding
18/20
Instruction
14/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate 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