qwen/qwen3.8-27b
LM-STUDIO MLXqwen3_5 · 27B · 4bit
Aug 19, 2026 · Apple M4 Max
deepseek-r1:1.5b
OLLAMA GGUFqwen2 · 1.8B · Q4_K_M
Jun 26, 2026 · Cortex-A78AE
Global Score
84 vs 60
Hardware Fit
69 vs 90
Quality Score
90 vs 47
Hardware
qwen/qwen3.8-27b deepseek-r1:1.5b
MachineMacBook ProNVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit Super
CPUApple M4 MaxCortex-A78AE
Cores166
RAM128 GB7 GB
GPUApple M4 MaxIntegrated / Unknown
OSmacOS 26.5.2Ubuntu 22.04.5 LTS
Archarm64arm64
Power Modebalancedbalanced
Performance
qwen/qwen3.8-27b deepseek-r1:1.5b
Tokens/sec28.934.6
First chunk33 ms5314 ms
TTFT4.4 s5.3 s
Load time7.9 s5.1 s
Memory usage0.0 GB1.1 GB
Memory %0%15%
HW Fit Score Breakdown
qwen/qwen3.8-27b
Speed
33/50
TTFT
6/20
Memory
30/30
deepseek-r1:1.5b
Speed
50/50
TTFT
10/20
Memory
30/30
Quality
qwen/qwen3.8-27b
Reasoning
19/20
Coding
15/20
Instruction
16/20
Structured
15/15
Math
15/15
Multilingual
10/10
Reasoning: Strong Coding: Adequate Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
deepseek-r1:1.5b
Reasoning
11/20
Coding
6/20
Instruction
7/20
Structured
4/15
Math
14/15
Multilingual
5/10
Reasoning: Adequate Coding: Weak Instruction Following: Weak Structured Output: Weak Math: Strong 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