gemma3:4b
OLLAMA GGUFgemma3 · 4.3B · Q4_K_M
Jun 25, 2026 · Cortex-A78AE
gemma4:26b
OLLAMA GGUFgemma4 · 25.2B · Q4_K_M
Oct 2, 2026 · AMD Ryzen 9 5950X 16-Core Processor
Global Score
46 vs 84
Hardware Fit
93 vs 80
Quality Score
26 vs 86
Hardware
gemma3:4b gemma4:26b
MachineNVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit SuperSystem manufacturer
CPUCortex-A78AEAMD Ryzen 9 5950X 16-Core Processor
Cores632
RAM7 GB32 GB
GPUIntegrated / UnknownNVIDIA GeForce RTX 3090
OSUbuntu 22.04.5 LTSMicrosoft Windows 11 Pro 10.0.26300
Archarm64x64
Power Modebalancedbalanced
Performance
gemma3:4b gemma4:26b
Tokens/sec14.9147.6
First chunk2204 ms139 ms
TTFT2.2 s30.0 s
Load time2.0 s21.3 s
Memory usage2.6 GB1.7 GB
Memory %35%5%
HW Fit Score Breakdown
gemma3:4b
Speed
45/50
TTFT
19/20
Memory
29/30
gemma4:26b
Speed
50/50
TTFT
0/20
Memory
30/30
Quality
gemma3:4b
Reasoning
11/20
Coding
11/20
Instruction
0/20
Structured
0/15
Math
4/15
Multilingual
0/10
Reasoning: Adequate Coding: Adequate Instruction Following: Poor Structured Output: Poor Math: Weak Multilingual: Poor
gemma4:26b
Reasoning
17/20
Coding
14/20
Instruction
18/20
Structured
15/15
Math
13/15
Multilingual
9/10
Reasoning: Strong Coding: Adequate Instruction Following: Strong 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