can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA GB300 Grace Blackwell (288GB HBM + 480GB) run DeepSeek V4 Flash (MoE)?

DeepSeek · MIT · estimated for a 288.0 GB machine at 4K context.

Yes — it runs great.

Best fit: DeepSeek V4 Flash (MoE) at Q3_K_M · 122.6 GB · 268–402 tok/s on a NVIDIA GB300 Grace Blackwell (288GB HBM + 480GB).

Every quantization on this machine

QuantSizeSpeedFit
Q3_K_M122.6 GB268–402 tok/sRuns great

Ran DeepSeek V4 Flash (MoE) on a NVIDIA GB300 Grace Blackwell (288GB HBM + 480GB)? Add your real numbers →