It hits the token limit.

Resolved 💬 2 comments Opened Nov 16, 2025 by EnterOgawa Closed Nov 26, 2025

What version of Codex is running?

0.58.00

What subscription do you have?

pro

Which model were you using?

gpt-5.1 high

What platform is your computer?

windows11pro

What issue are you seeing?

The following message appears and the session cannot continue.

■ Conversation is still above the token limit after automatic summarization (limit 244800, current 248148). Please start
a new session or trim your input.
■ stream disconnected before completion: stream closed before response.completed

<img width="635" height="70" alt="Image" src="https://github.com/user-attachments/assets/cdab2974-3fc6-41f7-b58d-8deecd9dc300" />

What steps can reproduce the bug?

When using it near the context limit, it ends up in this kind of state.
By the way, this session has repeatedly gone from full context to automatic compaction multiple times since 0.57.00.
I tried ending it once with Ctrl+C and resuming, but the situation didn’t change.
Even after using the /compact command, the situation remains unchanged.

What is the expected behavior?

_No response_

Additional information

_No response_

View original on GitHub ↗

This issue has 2 comments on GitHub. Read the full discussion on GitHub ↗