can-i-run-this-llm

Can I run this LLM?

Can a Desktop CPU, DDR5 dual-channel (32GB) run GLM-4 9B?

GLM · MIT · estimated for a 32.0 GB machine at 4K context.

Yes, but it's tight.

Best fit: GLM-4 9B at Q2_K · 4.0 GB · 7–10 tok/s on a Desktop CPU, DDR5 dual-channel (32GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K4.0 GB7–10 tok/sTight / slow
Q3_K_M5.0 GB6–9 tok/sTight / slow
Q4_K_M6.2 GB5–7 tok/sTight / slow
Q5_K_M7.1 GB4–6 tok/sTight / slow
Q6_K8.3 GB4–5 tok/sTight / slow
Q8_010.0 GB3–5 tok/sTight / slow
FP1618.8 GB2–2 tok/sTight / slow

Ran GLM-4 9B on a Desktop CPU, DDR5 dual-channel (32GB)? Add your real numbers →