Back to leaderboard

qwen3.5:4b

OLLAMA GGUF

qwen35 · 4.7B · Q4_K_M

Good

Aug 28, 2026 · Intel Core™ i5-10210U

llama3.2:3b

OLLAMA GGUF

llama · 3.2B · Q4_K_M

Good

Jun 25, 2026 · Cortex-A78AE

Global Score
64 vs 74
Hardware Fit
69 vs 99
Quality Score
62 vs 63

Hardware

qwen3.5:4b llama3.2:3b
MachineLENOVO 20U30029MBNVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit Super
CPUIntel Core™ i5-10210UCortex-A78AE
Cores86
RAM15 GB7 GB
GPUCometLake-U GT2 [UHD Graphics]Integrated / Unknown
OSPop 24.04 LTSUbuntu 22.04.5 LTS
Archx64arm64
Power Modeperformancebalanced

Performance

qwen3.5:4b llama3.2:3b
Tokens/sec7.421.1
First chunk1815 ms1032 ms
TTFT1.8 s1.0 s
Load time10.9 s8.4 s
Memory usage3.5 GB2.4 GB
Memory %23%32%

HW Fit Score Breakdown

qwen3.5:4b

Speed
19/50
TTFT
20/20
Memory
30/30

llama3.2:3b

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

Quality

qwen3.5:4b

Reasoning
15/20
Coding
14/20
Instruction
6/20
Structured
12/15
Math
6/15
Multilingual
9/10
Reasoning: Adequate Coding: Adequate Instruction Following: Weak Structured Output: Strong Math: Weak Multilingual: Strong

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