openai/gpt-oss-20b

gpt_oss · 20B · MXFP4

THINKING MODEL

Mac mini (Apple M4)

24 GB · macOS 26.3

Tested on April 1, 2026 · Submitted by cryptoepops

Top 15% Compare

Global Score

82 /100

Excellent

Hardware Fit

85/100

Quality

81/100

Get this model

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

MLX versions & conversions

Hardware

Machine: Mac mini
CPU: Apple M4
Cores: 10 total (4 perf + 6 eff)
Frequency: 2.4 GHz
RAM: 24 GB LPDDR5
GPU: Apple M4
OS: macOS 26.3
Arch: arm64
Power Mode: balanced

Performance

Tokens/sec Tokens generated per second — higher is better: 31.8
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±1.0
First chunk latency Delay before the first streamed chunk arrives from runtime: 32 ms
Time to first token How fast the model starts responding — lower is better: 8.4 s
Load time Time to load the model into memory before inference: 22.7 s
Memory usage RAM consumed during inference — percentage of total system memory: 0.0 GB (0%)
Total tokens Total number of tokens generated across all benchmark prompts: 1708
Thinking tokens (est.) Estimated internal reasoning tokens — models with "thinking" generate more tokens: ~950

Score breakdown

Speed

50/50

Time to first token

5/20

Memory

30/30

Quality

Reasoning

19/20

Coding

14/20

Instruction following

9/20

Structured output

14/15

Math

15/15

Multilingual

10/10

Category levels

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

Metadata

Spec version: 0.2.1
Runtime: LM Studio 0.4.8+1
Model format: MLX
Hardware profile: ENTRY
Result hash: ec5aecb1f1d586c1c19a66386c449d7e327b135ad659a7d2582f962780a7acb9

Interpretation

Hardware fit: 85/100. Overall suitability: EXCELLENT (Global 82/100). Category profile: Reasoning: Strong, Coding: Adequate, Instruction Following: Weak, Structured Output: Strong, Math: Strong, Multilingual: Strong.

Warnings

Significant swap activity during benchmark (+1.6 GB). Model may exceed available RAM — results are severely degraded.

Bench Environment

Power: AC Swap delta: +1.6 GB CPU load: avg 28% (peak 38%)

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