gemma4:31b
OLLAMA GGUFgemma4 · 30.7B · Q4_K_M
Oct 2, 2026 · AMD Ryzen 9 5950X 16-Core Processor
qwen3.5:27b
OLLAMA GGUFqwen35 · 27.8B · Q4_K_M
Mar 19, 2026 · Apple M4
Global Score
64 vs 66
Hardware Fit
41 vs 34
Quality Score
74 vs 79
Hardware
gemma4:31b qwen3.5:27b
MachineSystem manufacturerMac mini
CPUAMD Ryzen 9 5950X 16-Core ProcessorApple M4
Cores3210
RAM32 GB24 GB
GPUNVIDIA GeForce RTX 3090Apple M4
OSMicrosoft Windows 11 Pro 10.0.26300macOS 26.3
Archx64arm64
Power Modebalancedbalanced
Performance
gemma4:31b qwen3.5:27b
Tokens/sec9.53.8
First chunk1751 ms1637 ms
TTFT30.0 s1.6 s
Load time33.9 s17.0 s
Memory usage1.4 GB21.3 GB
Memory %4%89%
HW Fit Score Breakdown
gemma4:31b
Speed
11/50
TTFT
0/20
Memory
30/30
qwen3.5:27b
Speed
8/50
TTFT
20/20
Memory
6/30
Quality
gemma4:31b
Reasoning
11/20
Coding
17/20
Instruction
10/20
Structured
15/15
Math
12/15
Multilingual
9/10
Reasoning: Adequate Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Strong Multilingual: Strong
qwen3.5:27b
Reasoning
16/20
Coding
18/20
Instruction
8/20
Structured
15/15
Math
12/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Weak 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