gemma3:270m

gemma3 · 268.10M · Q8_0

MacBook Pro (Apple M3 Pro)

36 GB · macOS 15.7.5

Tested on March 5, 2026

Top 81% Compare

Global Score

50 /100

Marginal

Hardware Fit

97/100

Quality

30/100

Get this model

🦙

Ollama

ollama pull gemma3:270m

View on Ollama Library

ollama.com/library/gemma3

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 M3 Pro
Cores: 12 total (6 perf + 6 eff)
Frequency: 2.4 GHz
RAM: 36 GB LPDDR5
GPU: Apple M3 Pro
OS: macOS 15.7.5
Arch: arm64
Power Mode: balanced

Performance

Tokens/sec Tokens generated per second — higher is better: 239.2
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±12.1
First chunk latency Delay before the first streamed chunk arrives from runtime: 127 ms
Time to first token How fast the model starts responding — lower is better: 127 ms
Load time Time to load the model into memory before inference: 0.7 s
Memory usage RAM consumed during inference — percentage of total system memory: 0.6 GB (2%)
Total tokens Total number of tokens generated across all benchmark prompts: 814

Score breakdown

Speed

50/50

Time to first token

20/20

Memory

27/30

Quality

Reasoning

4/20

Coding

2/20

Instruction following

10/20

Structured output

8/15

Math

1/15

Multilingual

5/10

Category levels

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

Metadata

Spec version: 0.2.1
Runtime: Ollama 0.17.6
Model format: GGUF
Hardware profile: BALANCED
Result hash: 40f3aaec2c4d0d65447b84d41514428e68a486ac2e565d95c1dff279512089a8

Interpretation

Hardware fit: 97/100. Overall suitability: MARGINAL (Global 50/100). Category profile: Reasoning: Poor, Coding: Poor, Instruction Following: Adequate, Structured Output: Adequate, Math: Poor, Multilingual: Weak.

Bench Environment

Power: AC CPU load: avg 16% (peak 21%)

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