// RUN โ€” FITS

Can the NVIDIA RTX 6000 Ada 48GB run RWKV-6 (Finch) 3B?

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

RWKV-6 (Finch) 3B on the NVIDIA RTX 6000 Ada 48GB

QuantFits?Max contextSpeedBeyond context
FP16 yes 556k 113-134.7 t/s ~4.7-5.6 (slow)
Q8 yes 604k 226.1-269.4 t/s ~9.4-11.2 (slow)
Q4 yes 624k 403.7-481.1 t/s ~16.8-20 (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 6000 Ada 48GB run RWKV-6 (Finch) 3B?โ–ถ

Yes. At Q4 it generates ~403.7-481.1 tok/s and fits up to 624k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~16.8-20 tok/s.

How much context fits?โ–ถ

At Q4, up to about 624k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.