can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA L40S (48GB) run GLM-4 9B?

GLM · MIT · estimated for a 48.0 GB machine at 4K context.

Yes — it runs great.

Best fit: GLM-4 9B at FP16 · 18.8 GB · 31–46 tok/s on a NVIDIA L40S (48GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K3.1 GB129–193 tok/sRuns great
Q3_K_M4.1 GB107–161 tok/sRuns great
Q4_K_M5.7 GB84–126 tok/sRuns great
Q5_K_M6.7 GB74–111 tok/sRuns great
Q6_K7.8 GB66–99 tok/sRuns great
Q8_010.0 GB54–80 tok/sRuns great
FP1618.8 GB31–46 tok/sRuns great

Ran GLM-4 9B on a NVIDIA L40S (48GB)? Add your real numbers →