Choose model and effort at the start of a session and stay with them.
Switching models drops the cache, and switching mid-task makes the next turns slower and dearer.
The habits that make the difference. Tick the ones you already keep and see what is left.
Switching models drops the cache, and switching mid-task makes the next turns slower and dearer.
The first enable reads the whole conversation uncached.
/fast
/rewind to abandoning a path.It returns to an already cached prefix, which compaction does not.
/rewind