can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA GB300 Grace Blackwell (288GB HBM + 480GB) run Ornith 1.5 9B?

Ornith · MIT · estimated for a 288.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Ornith 1.5 9B at FP16 · 17.9 GB · 173–259 tok/s on a NVIDIA GB300 Grace Blackwell (288GB HBM + 480GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K4.1 GB291–437 tok/sRuns great
Q3_K_M4.9 GB279–419 tok/sRuns great
Q4_K_M5.9 GB267–400 tok/sRuns great
Q5_K_M6.9 GB256–384 tok/sRuns great
Q6_K7.7 GB247–370 tok/sRuns great
Q8_09.6 GB229–344 tok/sRuns great
FP1617.9 GB173–259 tok/sRuns great

Ran Ornith 1.5 9B on a NVIDIA GB300 Grace Blackwell (288GB HBM + 480GB)? Add your real numbers →