// RUN โ€” FITS

Can the AMD Radeon RX 7900 XTX run OpenCLIP ViT-H-14-378-quickgelu?

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

OpenCLIP ViT-H-14-378-quickgelu on the AMD Radeon RX 7900 XTX

QuantFits?Max contextSpeedBeyond context
FP16 yes 284k 306.5-374.6 t/s ~12.8-15.6 (slow)
Q8 yes 299k 612.9-749.2 t/s ~25.5-31.2 (slow)
Q4 yes 306k 1094.5-1337.8 t/s ~45.6-55.7 (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 OpenCLIP ViT-H-14-378-quickgelu?โ–ถ

Yes. At Q4 it generates ~1094.5-1337.8 tok/s and fits up to 306k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~45.6-55.7 tok/s.

How much context fits?โ–ถ

At Q4, up to about 306k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.