// RUN โ€” FITS

Can the NVIDIA RTX PRO 6000 Blackwell run DistilProtBert?

Yes โ€” here's how fast and up to what context, with real measured numbers.

DistilProtBert on the NVIDIA RTX PRO 6000 Blackwell

QuantFits?Max contextSpeedBeyond context
FP16 yes 1318k 2843.8-3389.2 t/s ~63.5-75.7 (slow)
Q8 yes 1321k 5687.7-6778.4 t/s ~127-151.3 (slow)
Q4 yes 1323k 10156.5-12104.3 t/s ~226.7-270.2 (slow)

Estimate (memory-bound). Beyond "max context" the KV cache spills to system RAM offload (depends on your system RAM) and speed drops to the "beyond" figure.

FAQ

Can the NVIDIA RTX PRO 6000 Blackwell run DistilProtBert?โ–ถ

Yes. At Q4 it generates ~10156.5-12104.3 tok/s and fits up to 1323k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~226.7-270.2 tok/s.

How much context fits?โ–ถ

At Q4, up to about 1323k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.