can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA DGX Spark (GB10, 128GB) run Granite 4.2 8B?

Granite · Apache 2.0 · estimated for a 128.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Granite 4.2 8B at Q8_0 · 9.4 GB · 17–25 tok/s on a NVIDIA DGX Spark (GB10, 128GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K3.4 GB36–54 tok/sRuns great
Q3_K_M4.4 GB30–45 tok/sRuns great
Q4_K_M5.4 GB26–39 tok/sRuns great
Q5_K_M6.3 GB23–35 tok/sRuns great
Q6_K7.2 GB21–31 tok/sRuns great
Q8_09.4 GB17–25 tok/sRuns great
FP1617.6 GB10–14 tok/sTight / slow

Ran Granite 4.2 8B on a NVIDIA DGX Spark (GB10, 128GB)? Add your real numbers →