// RUN โ€” FITS

Can the AMD Radeon RX 7900 XT run Mamba2-Hybrid?

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

Mamba2-Hybrid on the AMD Radeon RX 7900 XT

QuantFits?Max contextSpeedBeyond context
FP16 no โ€” โ€” โ€”
Q8 yes 63k 58.2-71.1 t/s ~2.9-3.6 (slow)
Q4 yes 92k 103.9-127 t/s ~5.2-6.4 (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 AMD Radeon RX 7900 XT run Mamba2-Hybrid?โ–ถ

Yes. At Q4 it generates ~103.9-127 tok/s and fits up to 92k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~5.2-6.4 tok/s.

How much context fits?โ–ถ

At Q4, up to about 92k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.