# Dex Horthy says compaction is now good enough to keep threads alive longer

> Original title: compaction is much better now so for certain things you may not need to explicitly clear or externalize context to for the sake of model ...

- Source: Dex Horthy
- Published: 2026-11-07T20:29:06.000Z
- Bad Influence: https://badinfluence.news/items/sqsr4mbjx8hafi18386et3qzm
- Original: https://x.com/dexhorthy/status/2119165267368698188

## Summary

Dex Horthy says compaction has improved enough that for some tasks users no longer need to clear or externalize context to keep models focused. He says he sets autocompact to 400k and runs manual compacts between 200k and 400k tokens, depending on the model and harness. He adds that three-level subagent architectures help, since a top-level orchestrator can launch subagents that launch their own.

## Post

compaction is much better now so for certain things you may not need to explicitly clear or externalize context to for the sake of model intelligence/focus. (still useful for collab w/ humans or handoff between agents / machines though!)

"dumb zone" a little more flexible for 1m frontier models. I usually my autocompact to 400k and tend to do a manual compact between 200-400k depending on model/harness. less so clearing now

3-tier subagent architectures make this a little better too. You can have a top level orchestrator that runs the process, and can launch subagents that can themselves launch subagents. So instead of human steering two levels, top level can. 

i don't think there's any good reason to go beyond 3-levels of subagent nesting today for majority of tasks.

YMMV, as always these are guidelines / training wheels. your intuition as to whether model is on track or getting dumb is always the best source!

## Quoted post @stepheniezzi

> Are yall still hyper focused on keeping the context window clean (ie /clear between tasks) or are you beginning to trust compacting and keeping threads alive longer?
> 
> I used to be locked in on @mattpocockuk and @dexhorthy trying everything to keep context tight. Much less so now

https://x.com/stepheniezzi/status/2119163030126678047