qwen3.8:27b
OLLAMA GGUFqwen35 · 27.3B · Q4_K_M
Aug 15, 2026 · AMD RYZEN AI MAX+ 395
gemma4:26b
OLLAMA GGUFgemma4 · 25.2B · Q4_K_M
Oct 2, 2026 · AMD Ryzen 9 5950X 16-Core Processor
Global Score
80 vs 84
Hardware Fit
66 vs 80
Quality Score
86 vs 86
Hardware
qwen3.8:27b gemma4:26b
MachineAZW GTR ProSystem manufacturer
CPUAMD RYZEN AI MAX+ 395AMD Ryzen 9 5950X 16-Core Processor
Cores3232
RAM125 GB32 GB
GPURadeon 8060SNVIDIA GeForce RTX 3090
OSUbuntu 24.04.4 LTSMicrosoft Windows 11 Pro 10.0.26300
Archx64x64
Power Modelow-powerbalanced
Performance
qwen3.8:27b gemma4:26b
Tokens/sec18.4147.6
First chunk587 ms139 ms
TTFT587 ms30.0 s
Load time7.4 s21.3 s
Memory usage17.4 GB1.7 GB
Memory %14%5%
HW Fit Score Breakdown
qwen3.8:27b
Speed
17/50
TTFT
19/20
Memory
30/30
gemma4:26b
Speed
50/50
TTFT
0/20
Memory
30/30
Quality
qwen3.8:27b
Reasoning
17/20
Coding
18/20
Instruction
16/20
Structured
14/15
Math
11/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Adequate Multilingual: Strong
gemma4:26b
Reasoning
17/20
Coding
14/20
Instruction
18/20
Structured
15/15
Math
13/15
Multilingual
9/10
Reasoning: Strong Coding: Adequate 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