can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA H200 (141GB) run Step 3.7 Flash (MoE)?

Step · Apache 2.0 · estimated for a 141.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Step 3.7 Flash (MoE) at Q3_K_M · 91.8 GB · 236–354 tok/s on a NVIDIA H200 (141GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K71.2 GB256–383 tok/sRuns great
Q3_K_M91.8 GB236–354 tok/sRuns great
Q4_K_M121.6 GB213–319 tok/sTight / slow
Q5_K_M142.0 GB199–299 tok/sWon't fit
Q6_K171.8 GB183–274 tok/sWon't fit
Q8_0212.0 GB164–246 tok/sWon't fit

Ran Step 3.7 Flash (MoE) on a NVIDIA H200 (141GB)? Add your real numbers →