// RUN โ€” FITS

Can the FuriosaAI RNGD run KataGo?

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

KataGo on the FuriosaAI RNGD

QuantFits?Max contextSpeedBeyond context
FP16 yes 651k 195000-237000 t/s ~5200-6320 (slow)
Q8 yes 651k 390000-474000 t/s ~10400-12640 (slow)
Q4 yes 651k 696428.6-846428.6 t/s ~18571.4-22571.4 (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 FuriosaAI RNGD run KataGo?โ–ถ

Yes. At Q4 it generates ~696428.6-846428.6 tok/s and fits up to 651k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~18571.4-22571.4 tok/s.

How much context fits?โ–ถ

At Q4, up to about 651k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.