gpt-oss:20b
OLLAMA GGUFgptoss · 20.9B · MXFP4
Aug 18, 2026 · Intel Core™ Ultra X7 358H
codestral:22b
OLLAMA GGUFllama · 22.2B · Q4_0
Jun 26, 2026 · Cortex-X925
Global Score
81 vs 67
Hardware Fit
59 vs 67
Quality Score
91 vs 67
Hardware
gpt-oss:20b codestral:22b
MachineGMKtec NucBox EVO-T2SNVIDIA NVIDIA_DGX_Spark
CPUIntel Core™ Ultra X7 358HCortex-X925
Cores1620
RAM64 GB122 GB
GPUIntel(R) Arc(TM) B390 GPUDevice 2e12
OSMicrosoft Windows 11 Pro 10.0.26200Ubuntu 24.04.4 LTS
Archx64arm64
Power Modebalancedperformance
Performance
gpt-oss:20b codestral:22b
Tokens/sec19.117.2
First chunk1985 ms129 ms
TTFT11.1 s129 ms
Load time7.7 s11.6 s
Memory usage12.9 GB18.9 GB
Memory %20%16%
HW Fit Score Breakdown
gpt-oss:20b
Speed
26/50
TTFT
3/20
Memory
30/30
codestral:22b
Speed
17/50
TTFT
20/20
Memory
30/30
Quality
gpt-oss:20b
Reasoning
18/20
Coding
19/20
Instruction
14/20
Structured
15/15
Math
15/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Strong Multilingual: Strong
codestral:22b
Reasoning
11/20
Coding
17/20
Instruction
9/20
Structured
14/15
Math
6/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Weak 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