qwen3:8b
OLLAMA GGUFqwen3 · 8.2B · Q4_K_M
Mar 5, 2026 · Intel Core™ Ultra 9 285K
gemma3:12b
OLLAMA GGUFgemma3 · 12.2B · Q4_K_M
Mar 2, 2026 · Apple M4
Global Score
84 vs 80
Hardware Fit
97 vs 83
Quality Score
79 vs 78
Hardware
qwen3:8b gemma3:12b
MachineASUSMacBook Air
CPUIntel Core™ Ultra 9 285KApple M4
Cores2410
RAM47 GB32 GB
GPUNVIDIA GeForce RTX 5090, Intel(R) GraphicsApple M4
OSMicrosoft Windows 11 Pro 10.0.26200macOS 26.3
Archx64arm64
Power Modebalancedbalanced
Performance
qwen3:8b gemma3:12b
Tokens/sec208.012.7
First chunk82 msN/A
TTFT1.3 s560 ms
Load time1.5 s2.8 s
Memory usage9.6 GB11.0 GB
Memory %20%34%
HW Fit Score Breakdown
qwen3:8b
Speed
50/50
TTFT
18/20
Memory
29/30
gemma3:12b
Speed
23/50
TTFT
30/20
Memory
30/30
Quality
qwen3:8b
Reasoning
18/20
Coding
10/20
Instruction
14/20
Structured
15/15
Math
12/15
Multilingual
10/10
Reasoning: Strong Coding: Weak Instruction Following: Adequate Structured Output: Strong Math: Strong Multilingual: Strong
gemma3:12b
Reasoning
14/20
Coding
19/20
Instruction
13/20
Structured
15/15
Math
7/15
Multilingual
10/10
Reasoning: Adequate 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