can-i-run-this-llm

Can I run this LLM?

Can a Mac Studio M3 Ultra (512GB) run DeepSeek V4 Flash (MoE)?

DeepSeek · MIT · estimated for a 512.0 GB machine at 4K context.

Yes — it runs great.

Best fit: DeepSeek V4 Flash (MoE) at Q4_K_M · 155.1 GB · 54–81 tok/s on a Mac Studio M3 Ultra (512GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K96.8 GB72–107 tok/sRuns great
Q3_K_M128.1 GB61–91 tok/sRuns great
Q4_K_M155.1 GB54–81 tok/sRuns great

Ran DeepSeek V4 Flash (MoE) on a Mac Studio M3 Ultra (512GB)? Add your real numbers →