// RUN โ€” FITS

Can the AMD Radeon RX 7900 XTX run ViT-Huge/14?

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

ViT-Huge/14 on the AMD Radeon RX 7900 XTX

QuantFits?Max contextSpeedBeyond context
FP16 yes 295k 478.5-584.8 t/s ~19.9-24.4 (slow)
Q8 yes 304k 957-1169.6 t/s ~39.9-48.7 (slow)
Q4 yes 309k 1708.9-2088.6 t/s ~71.2-87 (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 XTX run ViT-Huge/14?โ–ถ

Yes. At Q4 it generates ~1708.9-2088.6 tok/s and fits up to 309k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~71.2-87 tok/s.

How much context fits?โ–ถ

At Q4, up to about 309k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.