You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Add the dynamicThinkingEffort experimental setting (off by default) that gates the dynamic thinking-effort feature set: the model adjusting its own reasoning effort per turn via a native tool, and the user adjusting it in-chat.
Part of the Dynamic Thinking Effort series (5 PRs). Master design issue: easonLiangWorldedtech#28.
PR protocol (all series PRs): draft PR until CI is fully green -> switch to ready -> CodeRabbit bot review addressed in full -> 100% test coverage on patch lines. Each PR <= 1000 diff lines, based on latest upstream main (stacked on preceding series PRs where noted; rebased to pure main once dependencies merge).
Stacking: PR 1/5 — standalone, based on main.
Scope
packages/types/src/experiment.ts — experiment id dynamicThinkingEffort
Superseded by the DTE-v2 400-line redo. Content re-lands from the verified union 27a2e97 (feat/dte-trial-all, branch retained) as units U1-U39. New tracking issue: easonLiangWorldedtech#35.
Activity
easonLiangWorldedtech commentedon Sep 4, 2026
Superseded by the DTE-v2 400-line redo. Content re-lands from the verified union 27a2e97 (feat/dte-trial-all, branch retained) as units U1-U39. New tracking issue: easonLiangWorldedtech#35.