qwen2.5:3b
OLLAMA GGUFqwen2 · 3.1B · Q4_K_M
Jun 26, 2026 · Cortex-A78AE
qwen3.5:4b
OLLAMA GGUFqwen35 · 4.7B · Q4_K_M
Mar 5, 2026 · Apple M1 Max
Global Score
77 vs 83
Hardware Fit
100 vs 100
Quality Score
67 vs 76
Hardware
qwen2.5:3b qwen3.5:4b
MachineNVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit SuperMacBook Pro
CPUCortex-A78AEApple M1 Max
Cores610
RAM7 GB32 GB
GPUIntegrated / UnknownApple M1 Max
OSUbuntu 22.04.5 LTSmacOS 26.3
Archarm64arm64
Power Modebalancedbalanced
Performance
qwen2.5:3b qwen3.5:4b
Tokens/sec21.433.8
First chunk1006 ms293 ms
TTFT1.0 s293 ms
Load time6.4 s2.5 s
Memory usage2.0 GB6.5 GB
Memory %27%20%
HW Fit Score Breakdown
qwen2.5:3b
Speed
50/50
TTFT
20/20
Memory
30/30
qwen3.5:4b
Speed
50/50
TTFT
20/20
Memory
30/30
Quality
qwen2.5:3b
Reasoning
12/20
Coding
14/20
Instruction
11/20
Structured
15/15
Math
6/15
Multilingual
9/10
Reasoning: Adequate Coding: Adequate Instruction Following: Adequate Structured Output: Strong Math: Weak Multilingual: Strong
qwen3.5:4b
Reasoning
16/20
Coding
14/20
Instruction
15/20
Structured
15/15
Math
7/15
Multilingual
9/10
Reasoning: Strong Coding: Adequate Instruction Following: Strong Structured Output: Strong Math: Weak 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