qwen2.5:1.5b
OLLAMA GGUFqwen2 · 1.5B · Q4_K_M
Mar 16, 2026 · Intel Core™ i7-4771
qwen3:4b
OLLAMA GGUFqwen3 · 4.0B · Q4_K_M
Jun 26, 2026 · Cortex-A78AE
Global Score
65 vs 32
Hardware Fit
92 vs 98
Quality Score
54 vs 4
Hardware
qwen2.5:1.5b qwen3:4b
MachineiMacNVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit Super
CPUIntel Core™ i7-4771Cortex-A78AE
Cores86
RAM32 GB7 GB
GPUNVIDIA GeForce GTX 780MIntegrated / Unknown
OSmacOS 15.7.3Ubuntu 22.04.5 LTS
Archx64arm64
Power Modebalancedbalanced
Performance
qwen2.5:1.5b qwen3:4b
Tokens/sec19.617.9
First chunk690 ms950 ms
TTFT690 ms950 ms
Load time21.4 s9.7 s
Memory usage1.0 GB3.0 GB
Memory %3%40%
HW Fit Score Breakdown
qwen2.5:1.5b
Speed
42/50
TTFT
20/20
Memory
30/30
qwen3:4b
Speed
50/50
TTFT
20/20
Memory
28/30
Quality
qwen2.5:1.5b
Reasoning
8/20
Coding
9/20
Instruction
12/20
Structured
13/15
Math
5/15
Multilingual
7/10
Reasoning: Weak Coding: Weak Instruction Following: Adequate Structured Output: Strong Math: Weak 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