gemma-3-270m-it-qat-mlx

gemma3_text · 270M · 4bit

MacBook Air (Apple M4)

32 GB · macOS 26.3

Tested on March 4, 2026

Top 89% Compare

Global Score

43 /100

Marginal

Hardware Fit

99/100

Quality

19/100

Get this model

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

MLX versions & conversions

Hardware

Machine: MacBook Air
CPU: Apple M4
Cores: 10 total (4 perf + 6 eff)
Frequency: 2.4 GHz
RAM: 32 GB LPDDR5
GPU: Apple M4
OS: macOS 26.3
Arch: arm64
Power Mode: balanced

Performance

Tokens/sec Tokens generated per second — higher is better: 344.2
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±39.2
First chunk latency Delay before the first streamed chunk arrives from runtime: 173 ms
Time to first token How fast the model starts responding — lower is better: 173 ms
Load time Time to load the model into memory before inference: N/A
Memory usage RAM consumed during inference — percentage of total system memory: 0.3 GB (1%)
Total tokens Total number of tokens generated across all benchmark prompts: 879

Score breakdown

Speed

50/50

Time to first token

20/20

Memory

29/30

Quality

Reasoning

4/20

Coding

0/20

Instruction following

9/20

Structured output

3/15

Math

1/15

Multilingual

2/10

Category levels

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

Metadata

Spec version: 0.2.1
Runtime: LM Studio 0.4.6+1
Model format: MLX
Hardware profile: BALANCED
Result hash: d2d744e5d7b7ed5ab56aaa2c392bb89478f2f0647edeb5019901c4f868129299

Interpretation

Hardware fit: 99/100. Overall suitability: MARGINAL (Global 43/100). Category profile: Reasoning: Poor, Coding: Poor, Instruction Following: Weak, Structured Output: Poor, Math: Poor, Multilingual: Poor.

Bench Environment

Power: AC CPU load: avg 28% (peak 43%)

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