QFN
QFN Users Disable Thinking Preservation: A New Efficiency-Cost Trade-off
A QFN user found no capability loss after disabling preserved thinking, raising questions about reliability, context use, and inference costs.
Back to home
QFN
2 articles tagged with this topic
Tongyi Qianwen
27B Is Already Enough — Tongyi's New Model Ends the 'Bigger Is Better' Story
On AMD Strix, Tongyi's newest model barely beats 27B in daily work — a 'good enough' signal more valuable than any benchmark for enterprise IT.