can-i-run-this-llm

Can I run this LLM?

Can a MacBook Air M1 (8GB) run Llama 3.2 3B?

Llama · Llama 3.2 Community · estimated for a 8.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Llama 3.2 3B at Q6_K · 2.6 GB · 14–20 tok/s on a MacBook Air M1 (8GB).

Every quantization on this machine

QuantSizeSpeedFit
Q4_K_M1.9 GB17–26 tok/sRuns great
Q5_K_M2.3 GB15–23 tok/sRuns great
Q6_K2.6 GB14–20 tok/sRuns great
Q8_03.4 GB11–17 tok/sTight / slow
FP166.4 GB6–10 tok/sWon't fit

Ran Llama 3.2 3B on a MacBook Air M1 (8GB)? Add your real numbers →