// RUN — FITS

Can the GMKtec EVO-X3 run ERNIE-4.5-0.3B?

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

ERNIE-4.5-0.3B on the GMKtec EVO-X3

QuantFits?Max contextSpeedBeyond context
FP16 yes 1724k 231.1-280.9 t/s ~9-11 (slow)
Q8 yes 1729k 462.2-561.8 t/s ~18.1-21.9 (slow)
Q4 yes 1732k 825.4-1003.2 t/s ~32.2-39.2 (slow)

Estimate (memory-bound). Beyond "max context" the KV cache spills to SSD swap and speed drops to the "beyond" figure.

→ All models on the GMKtec EVO-X3 → ERNIE-4.5-0.3B specs → Other hardware

FAQ

Can the GMKtec EVO-X3 run ERNIE-4.5-0.3B?▶

Yes. At Q4 it generates ~825.4-1003.2 tok/s and fits up to 1732k context; beyond that it spills to SSD swap and slows to ~32.2-39.2 tok/s.

How much context fits?▶

At Q4, up to about 1732k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.