OpenAI tweaks Codex defaults after complaints over high-compute mode and desktop redesign

OpenAI says Codex will temporarily lift its five-hour cap for Plus, Business and Pro users and reset quotas once as it adjusts GPT-5.6 Sol usage calculations.

SOL

Summary

OpenAI has added temporary usage relief for Codex and ChatGPT Work after earlier complaints about confusing defaults and desktop design changes. Tibo said combined active users of Codex and ChatGPT Work have reached 6 million. Codex will temporarily remove its five-hour usage limit for Plus, Business and Pro plans and reset quotas once while OpenAI adjusts GPT-5.6 Sol usage calculations. The company had previously changed Codex defaults and its model selector after users said high-compute mode was too easy to enable without a clear warning on how it would affect limits, and Tibo also said a major desktop redesign made chat and project features harder to find.

Terms & Concepts
  • high-compute mode: A setting that uses more computing resources and can affect how quickly usage allowances are consumed.
  • usage limit: A cap on how much of a service a user can access over a given period.