granite4.2:3b

granite · 3.7B · Q4_K_M

MacBook Pro (Apple M1 Pro)

16 GB · macOS 26.6.2

Tested on August 29, 2026 · Submitted by AAA

Top 81% Compare

Global Score

59 /100

Marginal

Hardware Fit

80/100

Quality

50/100

Get this model

🦙

Ollama

ollama pull granite4.2:3b

View on Ollama Library

ollama.com/library/granite4.2

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 M1 Pro
Cores: 10 total (8 perf + 2 eff)
Frequency: 2.4 GHz
RAM: 16 GB LPDDR5
GPU: Apple M1 Pro
OS: macOS 26.6.2
Arch: arm64
Power Mode: balanced

Performance

Tokens/sec Tokens generated per second — higher is better: 46.3
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±0.6
First chunk latency Delay before the first streamed chunk arrives from runtime: 167 ms
Time to first token How fast the model starts responding — lower is better: 167 ms
Load time Time to load the model into memory before inference: 6.3 s
Memory usage RAM consumed during inference — percentage of total system memory: 12.7 GB (79%)
Total tokens Total number of tokens generated across all benchmark prompts: 1274

Score breakdown

Speed

50/50

Time to first token

20/20

Memory

10/30

Quality

Reasoning

18/20

Coding

10/20

Instruction following

5/20

Structured output

0/15

Math

15/15

Multilingual

2/10

Category levels

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

Metadata

Spec version: 0.2.1
Runtime: Ollama 0.33.1
Model format: GGUF
Hardware profile: ENTRY
Result hash: 1d8cf9c5f1379bfcffbfdc4fe9cd83606eb39f0f4210b67bd513540b9724689d

Interpretation

Hardware fit: 80/100. Overall suitability: MARGINAL (Global 59/100). Category profile: Reasoning: Strong, Coding: Adequate, Instruction Following: Weak, Structured Output: Poor, Math: Strong, Multilingual: Poor.

Warnings

Significant swap activity during benchmark (+1.1 GB). Model may exceed available RAM — results are severely degraded.
Running on battery power — performance may be reduced.

Bench Environment

Power: Battery Swap delta: +1.1 GB CPU load: avg 54% (peak 67%)

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