llama3:latest

llama · 8.0B · Q4_0

MacBook Air (Apple M4)

32 GB · macOS 26.3

Tested on March 2, 2026

Top 25% Compare

Global Score

74 /100

Good

Hardware Fit

95/100

Quality

60/100

Get this model

🦙

Ollama

ollama pull llama3:latest

View on Ollama Library

ollama.com/library/llama3

🤗

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: 22.3
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±0.1
Time to first token How fast the model starts responding — lower is better: 295 ms
Load time Time to load the model into memory before inference: 0.8 s
Memory usage RAM consumed during inference — percentage of total system memory: 5.6 GB (18%)
Total tokens Total number of tokens generated across all benchmark prompts: 1006

Score breakdown

Speed

35/40

Time to first token

30/30

Memory

30/30

Quality

Reasoning

9/20

Coding

9/20

Instruction following

15/20

Structured output

13/15

Math

4/15

Multilingual

10/10

Category levels

Reasoning: Weak Coding: Weak 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: fbde0cae382cc7cb58a0d159571ddb900a5882449807c019351026163125fd5d

Interpretation

Hardware fit: 95/100. Overall suitability: GOOD (Global 74/100). Category profile: Reasoning: Weak, Coding: Weak, Instruction Following: Strong, Structured Output: Strong, Math: Weak, Multilingual: Strong.

Run yours now

npx metrillm@latest bench

Requires Node 20+ and Ollama running