qwen3:32b
OLLAMA GGUFqwen3 · 32.8B · Q4_K_M
Jun 26, 2026 · Cortex-X925
gemma4:12b-mlx
OLLAMA GGUFAug 24, 2026 · Apple M1 Pro
Global Score
79 vs 71
Hardware Fit
59 vs 76
Quality Score
88 vs 69
Hardware
qwen3:32b gemma4:12b-mlx
MachineNVIDIA NVIDIA_DGX_SparkMacBook Pro
CPUCortex-X925Apple M1 Pro
Cores2010
RAM122 GB16 GB
GPUDevice 2e12Apple M1 Pro
OSUbuntu 24.04.4 LTSmacOS 26.6.2
Archarm64arm64
Power Modeperformancebalanced
Performance
qwen3:32b gemma4:12b-mlx
Tokens/sec9.725.0
First chunk324 ms787 ms
TTFT324 ms30.0 s
Load time24.5 s4.0 s
Memory usage26.8 GB7.1 GB
Memory %22%45%
HW Fit Score Breakdown
qwen3:32b
Speed
9/50
TTFT
20/20
Memory
30/30
gemma4:12b-mlx
Speed
50/50
TTFT
0/20
Memory
26/30
Quality
qwen3:32b
Reasoning
19/20
Coding
18/20
Instruction
15/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
gemma4:12b-mlx
Reasoning
14/20
Coding
10/20
Instruction
9/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Adequate Coding: Adequate Instruction Following: Weak Structured Output: Strong Math: Adequate 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