OpenWeightsTerminal

Qwen3.8-Flash-Next

Qwen3.8-Flash-Next

2× RTX 5060 Ti 16GB · IQ3_S · Strata

Decode · tok/s
99.9
Prefill
—
tok/100W
27.8
Testing Qwen 3.8 Flash IQ3_S (GSQ RCO) on 2× RTX 5060 Ti 16GB with Strata. The Monitor also recorded a separate completed request at 99.9 tok/s.

First-person pair of 16GB cards. Rank is the quoted 99.9 completed request. ~91.7 trace and ~2,866 prefill are tildes.

Officialclaimed
Open pair