winnow-12b
LM-STUDIO GGUFgemma4 · 12B · Q8_0
Nov 1, 2026 · AMD RYZEN AI MAX+ 395
Kimi-48B:48B
OLLAMA GGUFkimi-linear · 49.1B · unknown
Aug 19, 2026 · Intel Core™ Ultra X7 358H
Global Score
77 vs 86
Hardware Fit
38 vs 78
Quality Score
93 vs 89
Hardware
winnow-12b Kimi-48B:48B
MachineGMKtec EVO-X3GMKtec NucBox EVO-T2S
CPUAMD RYZEN AI MAX+ 395Intel Core™ Ultra X7 358H
Cores3216
RAM120 GB64 GB
GPUAMD Radeon(TM) 8060S GraphicsIntel(R) Arc(TM) B390 GPU
OSMicrosoft Windows 11 Pro 10.0.26300Microsoft Windows 11 Pro 10.0.26200
Archx64x64
Power Modebalancedbalanced
Performance
winnow-12b Kimi-48B:48B
Tokens/sec12.922.8
First chunk19 ms721 ms
TTFT9.7 s721 ms
Load time86.2 s23.0 s
Memory usage51.3 GB25.5 GB
Memory %43%40%
HW Fit Score Breakdown
winnow-12b
Speed
11/50
TTFT
0/20
Memory
27/30
Kimi-48B:48B
Speed
31/50
TTFT
20/20
Memory
27/30
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
Kimi-48B:48B
Reasoning
18/20
Coding
19/20
Instruction
15/20
Structured
15/15
Math
13/15
Multilingual
9/10
Reasoning: Strong Coding: Strong Instruction Following: Strong 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