can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA RTX Pro 6000 Blackwell (96GB) run Granite 4.2 30B?

Granite · Apache 2.0 · estimated for a 96.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Granite 4.2 30B at FP16 · 58.6 GB · 21–31 tok/s on a NVIDIA RTX Pro 6000 Blackwell (96GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K10.9 GB85–128 tok/sRuns great
Q3_K_M14.1 GB70–106 tok/sRuns great
Q4_K_M17.7 GB59–89 tok/sRuns great
Q5_K_M20.8 GB52–78 tok/sRuns great
Q6_K24.0 GB46–69 tok/sRuns great
Q8_031.1 GB37–55 tok/sRuns great
FP1658.6 GB21–31 tok/sRuns great

Ran Granite 4.2 30B on a NVIDIA RTX Pro 6000 Blackwell (96GB)? Add your real numbers →