GPT-20B:20b
OLLAMA GGUFgptoss · 20.9B · MXFP4
Aug 19, 2026 · Intel Core™ Ultra X7 358H
qwen3.8:latest
OLLAMA GGUFqwen35 · 27.3B · Q4_K_M
Oct 2, 2026 · AMD Ryzen 9 5950X 16-Core Processor
Global Score
79 vs 68
Hardware Fit
59 vs 22
Quality Score
88 vs 88
Hardware
GPT-20B:20b qwen3.8:latest
MachineGMKtec NucBox EVO-T2SSystem manufacturer
CPUIntel Core™ Ultra X7 358HAMD Ryzen 9 5950X 16-Core Processor
Cores1632
RAM64 GB32 GB
GPUIntel(R) Arc(TM) B390 GPUNVIDIA GeForce RTX 3090
OSMicrosoft Windows 11 Pro 10.0.26200Microsoft Windows 11 Pro 10.0.26300
Archx64x64
Power Modebalancedbalanced
Performance
GPT-20B:20b qwen3.8:latest
Tokens/sec19.310.0
First chunk2004 ms1041 ms
TTFT10.4 s14.3 s
Load time10.4 s22.3 s
Memory usage12.9 GB26.2 GB
Memory %20%82%
HW Fit Score Breakdown
GPT-20B:20b
Speed
26/50
TTFT
3/20
Memory
30/30
qwen3.8:latest
Speed
12/50
TTFT
1/20
Memory
9/30
Quality
GPT-20B:20b
Reasoning
19/20
Coding
18/20
Instruction
11/20
Structured
15/15
Math
15/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Strong Multilingual: Strong
qwen3.8:latest
Reasoning
19/20
Coding
19/20
Instruction
10/20
Structured
15/15
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