can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA H200 (141GB) run Llama 4 Scout (MoE)?

Llama · See model card · estimated for a 141.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Llama 4 Scout (MoE) at Q6_K · 88.4 GB · 145–218 tok/s on a NVIDIA H200 (141GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K39.6 GB212–318 tok/sRuns great
Q3_K_M51.8 GB190–285 tok/sRuns great
Q4_K_M65.4 GB171–256 tok/sRuns great
Q5_K_M76.6 GB157–236 tok/sRuns great
Q6_K88.4 GB145–218 tok/sRuns great
Q8_0114.5 GB124–186 tok/sTight / slow
FP16215.6 GB80–120 tok/sWon't fit

Ran Llama 4 Scout (MoE) on a NVIDIA H200 (141GB)? Add your real numbers →