TechFlow reports, on July 11, OpenAI Codex core member Tibo stated in a post that previously it was too easy for users to enable high-compute mode, but failed to fully explain its impact on usage limits, causing issues with some multi-agent workflows, plugin features, and other experiences, even leading some users to mistakenly believe Codex would be phased out in the future.
The first batch of improvement measures has been urgently launched, including adjusting default settings and the model selector to prevent users from mistakenly using high-cost configurations, and Codex and ChatGPT Work usage limits have been reset twice. The next update is expected to launch next week, conversation and project features will return to the sidebar in a more familiar, customizable form, while also enhancing the visibility of high-compute mode usage limits and reset times, further clarifying the respective use cases for ChatGPT Work and Codex.
