can-i-run-this-llm

Can I run this LLM?

Can a Mac Studio M3 Ultra (96GB) run North Mini Code 1.0 (MoE)?

North · Apache 2.0 · estimated for a 96.0 GB machine at 4K context.

Yes — it runs great.

Best fit: North Mini Code 1.0 (MoE) at Q8_0 · 32.4 GB · 81–121 tok/s on a Mac Studio M3 Ultra (96GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K10.5 GB119–178 tok/sRuns great
Q3_K_M14.2 GB110–165 tok/sRuns great
Q4_K_M19.2 GB100–150 tok/sRuns great
Q5_K_M23.0 GB94–141 tok/sRuns great
Q6_K25.5 GB90–135 tok/sRuns great
Q8_032.4 GB81–121 tok/sRuns great
FP1661.0 GB57–86 tok/sTight / slow

Ran North Mini Code 1.0 (MoE) on a Mac Studio M3 Ultra (96GB)? Add your real numbers →