Back to leaderboard

qwen2.5:72b

OLLAMA GGUF

qwen2 · 72.7B · Q4_K_M

Not Rec.

Jun 27, 2026 · Cortex-X925

qwen3:14b

OLLAMA GGUF

qwen3 · 14.8B · Q4_K_M

Excellent

Jun 13, 2026 · AMD Ryzen 9 3900X 12-Core Processor

Global Score
75 vs 82
Hardware Fit
50 vs 90
Quality Score
86 vs 79

Hardware

qwen2.5:72b qwen3:14b
MachineNVIDIA NVIDIA_DGX_SparkMicro-Star International Co., Ltd. MS-7C37
CPUCortex-X925AMD Ryzen 9 3900X 12-Core Processor
Cores2024
RAM122 GB64 GB
GPUDevice 2e12GameViewer Virtual Display Adapter, NVIDIA GeForce RTX 3080
OSUbuntu 24.04.4 LTSMicrosoft Windows 11 专业版 10.0.26200
Archarm64x64
Power Modeperformancebalanced

Performance

qwen2.5:72b qwen3:14b
Tokens/sec4.474.4
First chunk529 ms324 ms
TTFT529 ms2.6 s
Load time49.7 s5.1 s
Memory usage54.0 GB9.0 GB
Memory %44%14%

HW Fit Score Breakdown

qwen2.5:72b

Speed
4/50
TTFT
20/20
Memory
26/30

qwen3:14b

Speed
50/50
TTFT
10/20
Memory
30/30

Quality

qwen2.5:72b

Reasoning
19/20
Coding
19/20
Instruction
12/20
Structured
14/15
Math
12/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Strong Multilingual: Strong

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

Run yours and compare

$ npm install -g metrillm@latest
$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest