granite3.1-dense:8b

granite · 8.2B · Q4_K_M

MacBook Air (Apple M4)

32 GB · macOS 26.3

Tested on March 1, 2026

Top 17% Compare

Global Score

76 /100

Good

Hardware Fit

91/100

Quality

66/100

Get this model

🦙

Ollama

ollama pull granite3.1-dense:8b

View on Ollama Library

ollama.com/library/granite3.1-dense

🤗

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: 18.8
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±0.2
Time to first token How fast the model starts responding — lower is better: 308 ms
Load time Time to load the model into memory before inference: 0.6 s
Memory usage RAM consumed during inference — percentage of total system memory: 13.0 GB (41%)
Total tokens Total number of tokens generated across all benchmark prompts: 1270

Score breakdown

Speed

31/40

Time to first token

30/30

Memory

30/30

Quality

Reasoning

12/20

Coding

11/20

Instruction following

15/20

Structured output

14/15

Math

5/15

Multilingual

9/10

Category levels

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

Metadata

Spec version: 0.2.0
Runtime: Ollama 0.17.4
Model format: GGUF
Hardware profile: BALANCED
Result hash: c973db3dd1aac360253ad74d6838e531783de620f86be5521df22229b446a234

Interpretation

Hardware fit: 91/100. Overall suitability: GOOD (Global 76/100). Category profile: Reasoning: Adequate, Coding: Adequate, Instruction Following: Strong, Structured Output: Strong, Math: Weak, Multilingual: Strong.

Run yours now

npx metrillm@latest bench

Requires Node 20+ and Ollama running