[esperent]

> we preemptively trigger a summarization step and load that when the context-window fills up.

How does this differ from auto compact? Also, how do you prove that yours is better than using auto compact?

[ivzak]

For auto-compact, we do essentially the same Anthropic does, but at 85% filled context window. Then, when the window is 100% filled, we pull this precompaction + append accumulated 15%. This allows to run compaction instantly