qwen3:14b
OLLAMA GGUFqwen3 · 14.8B · Q4_K_M
Jun 13, 2026 · AMD Ryzen 9 3900X 12-Core Processor
gemma4:12b-it-qat
OLLAMA GGUFgemma4 · 11.9B · Q4_0
Jun 25, 2026 · Apple M4 Pro
Global Score
82 vs 88
Hardware Fit
90 vs 86
Quality Score
79 vs 89
Hardware
qwen3:14b gemma4:12b-it-qat
MachineMicro-Star International Co., Ltd. MS-7C37Mac mini
CPUAMD Ryzen 9 3900X 12-Core ProcessorApple M4 Pro
Cores2412
RAM64 GB24 GB
GPUGameViewer Virtual Display Adapter, NVIDIA GeForce RTX 3080Apple M4 Pro
OSMicrosoft Windows 11 专业版 10.0.26200macOS 15.6.1
Archx64arm64
Power Modebalancedbalanced
Performance
qwen3:14b gemma4:12b-it-qat
Tokens/sec74.430.0
First chunk324 ms686 ms
TTFT2.6 s7.1 s
Load time5.1 s5.4 s
Memory usage9.0 GB7.1 GB
Memory %14%30%
HW Fit Score Breakdown
qwen3:14b
Speed
50/50
TTFT
10/20
Memory
30/30
gemma4:12b-it-qat
Speed
50/50
TTFT
6/20
Memory
30/30
Quality
qwen3:14b
Reasoning
18/20
Coding
11/20
Instruction
17/20
Structured
14/15
Math
10/15
Multilingual
9/10
Reasoning: Strong Coding: Adequate Instruction Following: Strong Structured Output: Strong Math: Adequate Multilingual: Strong
gemma4:12b-it-qat
Reasoning
18/20
Coding
18/20
Instruction
14/20
Structured
15/15
Math
14/15
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