// RUN โ€” FITS

Can the FuriosaAI RNGD run Genie-SCOPe (bio)?

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

Genie-SCOPe (bio) on the FuriosaAI RNGD

QuantFits?Max contextSpeedBeyond context
FP16 yes 651k 118902.4-144512.2 t/s ~3170.7-3853.7 (slow)
Q8 yes 651k 237804.9-289024.4 t/s ~6341.5-7707.3 (slow)
Q4 yes 651k 424651.6-516115 t/s ~11324-13763.1 (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 Genie-SCOPe (bio)?โ–ถ

Yes. At Q4 it generates ~424651.6-516115 tok/s and fits up to 651k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~11324-13763.1 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.