can-i-run-this-llm

Can I run this LLM?

Can a MacBook Pro M2 Max (96GB) run GPT-OSS 120B (MoE)?

GPT-OSS · Apache 2.0 · estimated for a 96.0 GB machine at 4K context.

Yes, but it's tight.

Best fit: GPT-OSS 120B (MoE) at FP16 · 65.4 GB · 57–86 tok/s on a MacBook Pro M2 Max (96GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K62.6 GB59–88 tok/sTight / slow
Q3_K_M62.6 GB59–88 tok/sTight / slow
Q4_K_M62.8 GB58–88 tok/sTight / slow
Q5_K_M62.9 GB58–88 tok/sTight / slow
Q6_K63.3 GB58–87 tok/sTight / slow
Q8_063.4 GB58–87 tok/sTight / slow
FP1665.4 GB57–86 tok/sTight / slow

Ran GPT-OSS 120B (MoE) on a MacBook Pro M2 Max (96GB)? Add your real numbers →