hf.co/empero-ai/Qwen3.8-4B-Distill-GGUF:Q4_K_M

qwen35 · 4.33B · unknown

THINKING MODEL

MacBook Pro (Apple M4)

16 GB · macOS 26.6.1

Tested on August 28, 2026

Top 69% Compare

Global Score

66 /100

Good

Hardware Fit

100/100

Quality

52/100

Get this model

🦙

Ollama

ollama pull hf.co/empero-ai/Qwen3.8-4B-Distill-GGUF:Q4_K_M

View on Ollama Library

ollama.com/library/hf.co/empero-ai/Qwen3.8-4B-Distill-GGUF

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

GGUF versions & quantizations

Hardware

Machine: MacBook Pro
CPU: Apple M4
Cores: 10 total (4 perf + 6 eff)
Frequency: 2.4 GHz
RAM: 16 GB LPDDR5
GPU: Apple M4
OS: macOS 26.6.1
Arch: arm64
Power Mode: balanced

Performance

Tokens/sec Tokens generated per second — higher is better: 26.5
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±1.6
First chunk latency Delay before the first streamed chunk arrives from runtime: 447 ms
Time to first token How fast the model starts responding — lower is better: 447 ms
Load time Time to load the model into memory before inference: 3.3 s
Memory usage RAM consumed during inference — percentage of total system memory: 4.8 GB (30%)
Total tokens Total number of tokens generated across all benchmark prompts: 1429

Score breakdown

Speed

50/50

Time to first token

20/20

Memory

30/30

Quality

Reasoning

18/20

Coding

14/20

Instruction following

4/20

Structured output

1/15

Math

15/15

Multilingual

0/10

Category levels

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

Metadata

Spec version: 0.2.1
Runtime: Ollama 0.32.1
Model format: GGUF
Hardware profile: ENTRY
Result hash: 22876bb5b943faf7e079331c9c8ad3b6047cd81ae68e9f4220fcb078a8c9034a

Interpretation

Hardware fit: 100/100. Overall suitability: GOOD (Global 66/100). Category profile: Reasoning: Strong, Coding: Adequate, Instruction Following: Poor, Structured Output: Poor, Math: Strong, Multilingual: Poor.

Warnings

Significant swap activity during benchmark (+3.1 GB). Model may exceed available RAM — results are severely degraded.

Bench Environment

Power: AC Swap delta: +3.1 GB CPU load: avg 18% (peak 24%)

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