phi:2.7b

phi2 · 3B · Q4_0

MacBook Air (Apple M4)

32 GB · macOS 26.3

Tested on March 2, 2026

Top 60% Compare

Global Score

65 /100

Good

Hardware Fit

100/100

Quality

41/100

Get this model

🦙

Ollama

ollama pull phi:2.7b

View on Ollama Library

ollama.com/library/phi

🤗

Find on HuggingFace

GGUF versions & quantizations

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: 56.6
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±0.6
Time to first token How fast the model starts responding — lower is better: 110 ms
Load time Time to load the model into memory before inference: 0.4 s
Memory usage RAM consumed during inference — percentage of total system memory: 2.2 GB (7%)
Total tokens Total number of tokens generated across all benchmark prompts: 1227

Score breakdown

Speed

40/40

Time to first token

30/30

Memory

30/30

Quality

Reasoning

7/20

Coding

4/20

Instruction following

9/20

Structured output

11/15

Math

8/15

Multilingual

2/10

Category levels

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

Metadata

Spec version: 0.2.0
Runtime: Ollama 0.17.4
Model format: GGUF
Hardware profile: BALANCED
Result hash: fda0c2f1e5ce4276711516a759a03cc22c054b488e224f8bdd824925ce33918c

Interpretation

Hardware fit: 100/100. Overall suitability: GOOD (Global 65/100). Category profile: Reasoning: Weak, Coding: Poor, Instruction Following: Weak, Structured Output: Adequate, Math: Adequate, Multilingual: Poor.

Run yours now

npx metrillm@latest bench

Requires Node 20+ and Ollama running