Back to leaderboard

qwen3.5:4b

OLLAMA GGUF

qwen35 · 4.2B · Q4_K_M

Marginal

Oct 12, 2026 · AMD Ryzen 5 5500

llama3.2:3b

OLLAMA GGUF

llama · 3.2B · Q4_K_M

Good

Jun 25, 2026 · Cortex-A78AE

Global Score
50 vs 74
Hardware Fit
93 vs 99
Quality Score
31 vs 63

Hardware

qwen3.5:4b llama3.2:3b
MachineMicrosoft WSLNVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit Super
CPUAMD Ryzen 5 5500Cortex-A78AE
Cores126
RAM8 GB7 GB
GPUIntegrated / UnknownIntegrated / Unknown
OSUbuntu 26.04.1 LTSUbuntu 22.04.5 LTS
Archx64arm64
Power Modeunknownbalanced

Performance

qwen3.5:4b llama3.2:3b
Tokens/sec92.621.1
First chunk323 ms1032 ms
TTFT2.8 s1.0 s
Load time9.0 s8.4 s
Memory usage2.9 GB2.4 GB
Memory %38%32%

HW Fit Score Breakdown

qwen3.5:4b

Speed
50/50
TTFT
15/20
Memory
28/30

llama3.2:3b

Speed
50/50
TTFT
20/20
Memory
29/30

Quality

qwen3.5:4b

Reasoning
18/20
Coding
1/20
Instruction
0/20
Structured
0/15
Math
12/15
Multilingual
0/10
Reasoning: Strong Coding: Poor Instruction Following: Poor Structured Output: Poor Math: Strong Multilingual: Poor

llama3.2:3b

Reasoning
10/20
Coding
14/20
Instruction
15/20
Structured
12/15
Math
3/15
Multilingual
9/10
Reasoning: Adequate Coding: Adequate Instruction Following: Strong Structured Output: Strong Math: Poor Multilingual: Strong
View qwen3.5:4b details | View llama3.2:3b details

Run yours and compare

$ npm install -g metrillm@latest
$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest