// RUN โ€” FITS

Can the NVIDIA B200 run BlenderBot 3?

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

BlenderBot 3 on the NVIDIA B200

QuantFits?Max contextSpeedBeyond context
FP16 no โ€” โ€” โ€”
Q8 yes 267 33.4-39.8 t/s ~0.2-0.2 (slow)
Q4 yes 147k 59.6-71 t/s ~0.3-0.4 (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 B200 run BlenderBot 3?โ–ถ

Yes. At Q4 it generates ~59.6-71 tok/s and fits up to 147k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~0.3-0.4 tok/s.

How much context fits?โ–ถ

At Q4, up to about 147k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.