How to enable 1M-token context window in GPT-5.6 Sol

JeremyNguyenPhD · x · 2026-08-18

This post explains how to enable a 1-million-token context window for GPT-5.6 Sol within Codex. By updating the /.codex/config.toml file with specific settings for model, modelcontextwindow, and modelautocompacttokenlimit, users can allow the model to retain significantly more code and conversation history before summarization.

Related event: Codex Users Unlock GPT-5.6 Sol 1M Context Window, at Double Cost Beyond 272K Tokens(12 posts)→

Original post →

More from coding & agent

coding & agent channel →