spb/wp13_uqo

1\begin{table}[htbp]2\centering3\caption{User-selectable model tier. The orchestration service enables adaptive extended4thinking only for models that support it, and applies a uniform per-response generation cap.5The default is a reasoning-augmented model with a very large context window; alternatives let6users trade off capability, latency, and cost per query.}7\label{tab:models}8\begin{threeparttable}9\begin{tabular}{@{}llcc@{}}10\toprule11Role in platform & Model family & Adaptive thinking & Typical use \\12\midrule13Default            & Reasoning-augmented (1M ctx) & Yes & Comprehensive reports \\14High capability    & Flagship reasoning           & Yes & Hardest analyses \\15Balanced           & Mid-tier reasoning           & Yes & Moderate queries \\16Low latency        & Fast tier                    & No  & Simple lookups \\17\bottomrule18\end{tabular}19\begin{tablenotes}[flushleft]20\footnotesize21\item \textit{Notes.} Per-response generation is capped at 64{,}000 tokens. Adaptive thinking22lets the model allocate a variable budget of private reasoning before responding; its summary is23streamed to the client and shown in a collapsible panel. Context-window sizes and per-token24prices are determined by the model provider and abstracted away from the orchestration logic.25\end{tablenotes}26\end{threeparttable}27\end{table}28