granite4.2:3b
OLLAMA GGUFgranite · 3.7B · Q4_K_M
Aug 29, 2026 · Apple M1 Pro
GPT-20B:20b
OLLAMA GGUFgptoss · 20.9B · MXFP4
Aug 24, 2026 · Intel Core™ Ultra X7 358H
Global Score
59 vs 78
Hardware Fit
80 vs 59
Quality Score
50 vs 86
Hardware
granite4.2:3b GPT-20B:20b
MachineMacBook ProGMKtec NucBox EVO-T2S
CPUApple M1 ProIntel Core™ Ultra X7 358H
Cores1016
RAM16 GB64 GB
GPUApple M1 ProIntel(R) Arc(TM) B390 GPU
OSmacOS 26.6.2Microsoft Windows 11 Pro 10.0.26200
Archarm64x64
Power Modebalancedbalanced
Performance
granite4.2:3b GPT-20B:20b
Tokens/sec46.319.6
First chunk167 ms2068 ms
TTFT167 ms9.9 s
Load time6.3 s12.8 s
Memory usage12.7 GB12.9 GB
Memory %79%20%
HW Fit Score Breakdown
granite4.2:3b
Speed
50/50
TTFT
20/20
Memory
10/30
GPT-20B:20b
Speed
26/50
TTFT
3/20
Memory
30/30
Quality
granite4.2:3b
Reasoning
18/20
Coding
10/20
Instruction
5/20
Structured
0/15
Math
15/15
Multilingual
2/10
Reasoning: Strong Coding: Adequate Instruction Following: Weak Structured Output: Poor Math: Strong Multilingual: Poor
GPT-20B:20b
Reasoning
18/20
Coding
18/20
Instruction
11/20
Structured
14/15
Math
15/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Strong 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