// RUN โ€” FITS

Can the NVIDIA RTX PRO 6000 Blackwell run Generative BST?

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

Generative BST on the NVIDIA RTX PRO 6000 Blackwell

QuantFits?Max contextSpeedBeyond context
FP16 yes 518k 69.3-82.6 t/s ~1.5-1.8 (slow)
Q8 yes 590k 138.7-165.3 t/s ~3.1-3.7 (slow)
Q4 yes 622k 247.7-295.2 t/s ~5.5-6.6 (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 Generative BST?โ–ถ

Yes. At Q4 it generates ~247.7-295.2 tok/s and fits up to 622k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~5.5-6.6 tok/s.

How much context fits?โ–ถ

At Q4, up to about 622k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.