gemma4:26b
OLLAMA GGUFgemma4 · 25.8B · Q4_K_M
Apr 3, 2026 · Apple M4
starcoder2:3b
OLLAMA GGUFstarcoder2 · 3B · Q4_0
Jun 26, 2026 · Cortex-A78AE
Global Score
33 vs 35
Hardware Fit
50 vs 100
Quality Score
25 vs 7
Hardware
gemma4:26b starcoder2:3b
MachineMac miniNVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit Super
CPUApple M4Cortex-A78AE
Cores106
RAM24 GB7 GB
GPUApple M4Integrated / Unknown
OSmacOS 26.3Ubuntu 22.04.5 LTS
Archarm64arm64
Power Modebalancedbalanced
Performance
gemma4:26b starcoder2:3b
Tokens/sec20.031.5
First chunk941 ms392 ms
TTFT30.0 s392 ms
Load time0.2 s4.8 s
Memory usage20.3 GB1.8 GB
Memory %85%24%
HW Fit Score Breakdown
gemma4:26b
Speed
42/50
TTFT
0/20
Memory
8/30
starcoder2:3b
Speed
50/50
TTFT
20/20
Memory
30/30
Quality
gemma4:26b
Reasoning
14/20
Coding
1/20
Instruction
6/20
Structured
3/15
Math
0/15
Multilingual
1/10
Reasoning: Adequate Coding: Poor Instruction Following: Weak Structured Output: Poor Math: Poor Multilingual: Poor
starcoder2:3b
Reasoning
0/20
Coding
3/20
Instruction
1/20
Structured
0/15
Math
1/15
Multilingual
2/10
Reasoning: Poor Coding: Poor Instruction Following: Poor Structured Output: Poor Math: Poor Multilingual: Poor
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