gemma4:e4b
OLLAMA GGUFgemma4 · 8.0B · Q4_K_M
Apr 3, 2026 · AMD RYZEN AI MAX+ 395
starcoder2:3b
OLLAMA GGUFstarcoder2 · 3B · Q4_0
Mar 2, 2026 · Apple M4
Global Score
84 vs 39
Hardware Fit
98 vs 100
Quality Score
78 vs 13
Hardware
gemma4:e4b starcoder2:3b
MachineAZW GTR ProMacBook Air
CPUAMD RYZEN AI MAX+ 395Apple M4
Cores3210
RAM125 GB32 GB
GPURadeon 8060SApple M4
OSUbuntu 24.04.4 LTSmacOS 26.3
Archx64arm64
Power Modeperformancebalanced
Performance
gemma4:e4b starcoder2:3b
Tokens/sec53.049.4
First chunk214 msN/A
TTFT214 ms155 ms
Load time3.1 s0.4 s
Memory usage15.0 GB3.0 GB
Memory %12%9%
HW Fit Score Breakdown
gemma4:e4b
Speed
48/50
TTFT
20/20
Memory
30/30
starcoder2:3b
Speed
40/50
TTFT
30/20
Memory
30/30
Quality
gemma4:e4b
Reasoning
14/20
Coding
17/20
Instruction
15/20
Structured
15/15
Math
7/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Weak Multilingual: Strong
starcoder2:3b
Reasoning
0/20
Coding
2/20
Instruction
5/20
Structured
2/15
Math
2/15
Multilingual
2/10
Reasoning: Poor Coding: Poor Instruction Following: Weak 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