winnow-12b
LM-STUDIO GGUFgemma4 · 12B · Q8_0
Nov 1, 2026 · AMD RYZEN AI MAX+ 395
phi4:14b
OLLAMA GGUFphi3 · 14.7B · Q4_K_M
Jun 26, 2026 · Cortex-X925
Global Score
77 vs 77
Hardware Fit
38 vs 74
Quality Score
93 vs 78
Hardware
winnow-12b phi4:14b
MachineGMKtec EVO-X3NVIDIA NVIDIA_DGX_Spark
CPUAMD RYZEN AI MAX+ 395Cortex-X925
Cores3220
RAM120 GB122 GB
GPUAMD Radeon(TM) 8060S GraphicsDevice 2e12
OSMicrosoft Windows 11 Pro 10.0.26300Ubuntu 24.04.4 LTS
Archx64arm64
Power Modebalancedperformance
Performance
winnow-12b phi4:14b
Tokens/sec12.922.9
First chunk19 ms198 ms
TTFT9.7 s198 ms
Load time86.2 s10.3 s
Memory usage51.3 GB14.7 GB
Memory %43%12%
HW Fit Score Breakdown
winnow-12b
Speed
11/50
TTFT
0/20
Memory
27/30
phi4:14b
Speed
24/50
TTFT
20/20
Memory
30/30
Quality
winnow-12b
Reasoning
19/20
Coding
19/20
Instruction
18/20
Structured
15/15
Math
12/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
phi4:14b
Reasoning
18/20
Coding
18/20
Instruction
12/20
Structured
13/15
Math
7/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate 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