can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA GB300 Grace Blackwell (288GB HBM + 480GB) run Step 3.7 Flash (MoE)?

Step · Apache 2.0 · estimated for a 288.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Step 3.7 Flash (MoE) at Q8_0 · 212.0 GB · 210–316 tok/s on a NVIDIA GB300 Grace Blackwell (288GB HBM + 480GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K71.2 GB291–436 tok/sRuns great
Q3_K_M91.8 GB276–413 tok/sRuns great
Q4_K_M121.6 GB256–384 tok/sRuns great
Q5_K_M142.0 GB244–366 tok/sRuns great
Q6_K171.8 GB229–343 tok/sRuns great
Q8_0212.0 GB210–316 tok/sRuns great

Ran Step 3.7 Flash (MoE) on a NVIDIA GB300 Grace Blackwell (288GB HBM + 480GB)? Add your real numbers →