can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA RTX 3090 (24GB) run Ornith 1.5 35B-A3B (MoE)?

Ornith · MIT · estimated for a 24.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Ornith 1.5 35B-A3B (MoE) at Q3_K_M · 16.7 GB · 204–306 tok/s on a NVIDIA RTX 3090 (24GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K13.1 GB224–336 tok/sRuns great
Q3_K_M16.7 GB204–306 tok/sRuns great
Q4_K_M21.9 GB181–271 tok/sWon't fit
Q5_K_M25.5 GB167–251 tok/sWon't fit
Q6_K30.5 GB152–228 tok/sWon't fit
Q8_037.8 GB134–201 tok/sWon't fit
FP1671.1 GB87–130 tok/sWon't fit

Ran Ornith 1.5 35B-A3B (MoE) on a NVIDIA RTX 3090 (24GB)? Add your real numbers →