qwen3.5:0.8b

qwen35 · 752.39M · Q8_0

THINKING MODEL

Microsoft WSL (AMD Ryzen 5 5500)

8 GB · Ubuntu 26.04.1 LTS

Tested on October 11, 2026

Top 92% Compare

Global Score

44 /100

Not Rec.

Hardware Fit

80/100

Quality

28/100

Get this model

🦙

Ollama

ollama pull qwen3.5:0.8b

View on Ollama Library

ollama.com/library/qwen3.5

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

GGUF versions & quantizations

Hardware

Machine: Microsoft WSL
CPU: AMD Ryzen 5 5500
Cores: 12 threads (6 cores)
RAM: 8 GB
GPU: Integrated / Unknown
OS: Ubuntu 26.04.1 LTS
Arch: x64

Performance

Tokens/sec Tokens generated per second — higher is better: 126.1
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±4.9
First chunk latency Delay before the first streamed chunk arrives from runtime: 64 ms
Time to first token How fast the model starts responding — lower is better: 30.0 s
Load time Time to load the model into memory before inference: 0.0 s
Memory usage RAM consumed during inference — percentage of total system memory: 0.5 GB (7%)
Total tokens Total number of tokens generated across all benchmark prompts: 1429
Thinking tokens (est.) Estimated internal reasoning tokens — models with "thinking" generate more tokens: ~777

Score breakdown

Speed

50/50

Time to first token

0/20

Memory

30/30

Quality

Reasoning

9/20

Coding

0/20

Instruction following

7/20

Structured output

1/15

Math

6/15

Multilingual

5/10

Category levels

Reasoning: Weak Coding: Poor Instruction Following: Weak Structured Output: Poor Math: Weak Multilingual: Adequate

Metadata

Spec version: 0.2.1
Runtime: Ollama 0.34.0
Model format: GGUF
Hardware profile: ENTRY
Result hash: 6cbf1701a98cd7f0f35ce0d18a9aa1c9057bf236396e024b9339ad0fdee1a16d

Interpretation

Hardware fit: 80/100. Overall suitability: NOT RECOMMENDED (Global 44/100). Category profile: Reasoning: Weak, Coding: Poor, Instruction Following: Weak, Structured Output: Poor, Math: Weak, Multilingual: Adequate.

Disqualifiers

Time to first token too high: 30000ms (maximum: 22000ms for ENTRY profile)

Bench Environment

CPU load: avg 25% (peak 31%)

Run yours now

$ npm install -g metrillm@latest

$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest