GPT-5.5 Fast mode remains extremely slow in Codex 26.609.30741

Summary

On June 12, 2026, Codex is still extremely slow when using GPT-5.5 with Fast mode enabled. Even simple requests can take a very long time to respond, and sometimes the task appears unable to complete at all.

Environment

• Codex version: 26.609.30741
• Release shown in app: Jun 11, 2026
• Model: GPT-5.5
• Mode: Fast mode
• Date observed: June 12, 2026

Actual behavior

• Simple requests frequently require a very long wait before Codex responds.
• Codex often spends a long time thinking without producing a usable response.
• Some requests appear to get stuck or fail to complete.
• This continues to happen after updating to the latest available version, 26.609.30741.

Expected behavior

Fast mode should provide noticeably responsive behavior, especially for simple requests. Basic interactions should complete quickly enough for normal coding workflows.

Impact

The current response speed makes Codex very inefficient to use. In practice, it is difficult or even impossible to consume the purchased plan quota because requests take so long or fail to complete. This makes the plan feel much less useful than expected.

Additional context

This does not appear to be a one-off slow request. The slow thinking / no-response behavior happens repeatedly even after updating to the latest version.