// RUN โ€” FITS

Can the AMD Instinct MI455X run Routing Transformer (WT-103)?

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

Routing Transformer (WT-103) on the AMD Instinct MI455X

QuantFits?Max contextSpeedBeyond context
FP16 yes 6039k 77660.4-94918.2 t/s ~158.5-193.7 (slow)
Q8 yes 6040k 155320.8-189836.5 t/s ~317-387.4 (slow)
Q4 yes 6041k 277358.5-338993.7 t/s ~566-691.8 (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 Instinct MI455X run Routing Transformer (WT-103)?โ–ถ

Yes. At Q4 it generates ~277358.5-338993.7 tok/s and fits up to 6041k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~566-691.8 tok/s.

How much context fits?โ–ถ

At Q4, up to about 6041k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.