Codex Allows Raising Compaction Thresholds Per Directory
soumitrashukla9 · x · 2026-07-12
A follow-up explains a more granular use case: if you prefer not to change the default settings globally, you can place your automation tasks into a specific folder and add a project-level configuration file there.
This way, you only raise modelautocompacttokenlimit to 900000 for large tasks that need it, leaving the rest at the default 353K. The author emphasizes that the reason for this is straightforward: larger contexts are more expensive, so it should be enabled strictly on a per-need basis.
Related event: OpenAI Codex Raises Auto-Compaction Context Baseline(3 posts)→
More from coding & agent
- A dentist with a 3D printer discovered OpenAI Codex — and never looked back — gabrielchua · 2026-10-27
- AutoResearch agent produces weird 'forking' loss curves, and the team dug into the science — ZimingLiu11 · 2026-10-27
- Stanford's PaperFormalizer turns research papers into Lean-verified versions with author in the loop — StanfordUniversity · 2026-10-27
- Hugging Face TRL makes score centering + masked importance sampling default in async GRPO — _lewtun · 2026-10-27
- Firecrawl offers 2.5x credits to a16z Speedrun founders, unveils $75M Series B — devdigest · 2026-10-27
- Chums: open-source Apple Watch app puts Claude Code and Codex on your wrist — brandon_galang · 2026-10-27