llama3.2:1b
OLLAMA GGUFllama · 1.2B · Q8_0
Mar 5, 2026 · AMD Ryzen 9 8945HS
qwen3:4b
OLLAMA GGUFqwen3 · 4.0B · Q4_K_M
Jun 26, 2026 · Cortex-A78AE
Global Score
62 vs 29
Hardware Fit
91 vs 86
Quality Score
50 vs 4
Hardware
llama3.2:1b qwen3:4b
MachineASUSTeK COMPUTER INC. ROG Zephyrus G14 GA403UV_GA403UVNVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit Super
CPUAMD Ryzen 9 8945HSCortex-A78AE
Cores166
RAM15 GB7 GB
GPUNVIDIA GeForce RTX 4060 Laptop GPU, AMD Radeon 780M GraphicsIntegrated / Unknown
OSMicrosoft Windows 11 Home 10.0.26200Ubuntu 22.04.5 LTS
Archx64arm64
Power Modebalancedbalanced
Performance
llama3.2:1b qwen3:4b
Tokens/sec170.217.9
First chunk142 ms1010 ms
TTFT142 ms6.4 s
Load time3.2 s9.5 s
Memory usage1.6 GB3.0 GB
Memory %11%40%
HW Fit Score Breakdown
llama3.2:1b
Speed
50/50
TTFT
20/20
Memory
21/30
qwen3:4b
Speed
50/50
TTFT
8/20
Memory
28/30
Quality
llama3.2:1b
Reasoning
4/20
Coding
13/20
Instruction
14/20
Structured
13/15
Math
0/15
Multilingual
6/10
Reasoning: Poor Coding: Adequate Instruction Following: Adequate Structured Output: Strong Math: Poor Multilingual: Adequate
qwen3:4b
Reasoning
4/20
Coding
0/20
Instruction
0/20
Structured
0/15
Math
0/15
Multilingual
0/10
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