can-i-run-this-llm

Can I run this LLM?

Can a Desktop CPU, DDR5 dual-channel (32GB) run Ornith 1.5 35B-A3B (MoE)?

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

Yes — it runs great.

Best fit: Ornith 1.5 35B-A3B (MoE) at Q4_K_M · 21.9 GB · 13–20 tok/s on a Desktop CPU, DDR5 dual-channel (32GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K13.1 GB19–29 tok/sRuns great
Q3_K_M16.7 GB16–24 tok/sRuns great
Q4_K_M21.9 GB13–20 tok/sRuns great
Q5_K_M25.5 GB12–18 tok/sWon't fit
Q6_K30.5 GB10–15 tok/sWon't fit
Q8_037.8 GB9–13 tok/sWon't fit
FP1671.1 GB5–7 tok/sWon't fit

Ran Ornith 1.5 35B-A3B (MoE) on a Desktop CPU, DDR5 dual-channel (32GB)? Add your real numbers →