[Codex Quota Issues Confirmed: OpenAI Identifies 3 Sources of Extra Usage, Full Reset Tomorrow] Beating AI Newsflash: OpenAI has finally pinpointed the specific reasons behind the rapid depletion of Codex quotas. Team lead Tibo Sottiaux stated that the team has identified three sources of extra usage: inefficiencies in compressing context during repeated operations in ultra-long sessions with images, excessive usage by Computer History in certain high-consumption scenarios, and even the auto-generated conversation title feature consuming more quota than expected. Previously, Tibo had claimed no overall anomalies were found and directed some affected users to sub2api and subscription sharing. However, he later admitted that the cache hit rate for some users had indeed worsened, potentially accelerating quota consumption. Now, the investigation has finally pinpointed several specific issues within Codex itself. Among these, "context compression" is a mechanism designed to allow Codex to continuously compress old content during ultra-long tasks, freeing up context space to continue working. OpenAI's official documentation also confirms that Codex automatically uses compaction to sustain long-duration tasks. However, when there are many images and multiple rounds of compression, the current process results in additional waste. As for Computer History, it is a recently launched feature that brings operation records from selected apps and web pages on Mac into ChatGPT and Codex. OpenAI plans to roll out fixes on Sunday, U.S. time, and will perform a full reset of Codex quotas for all paid subscribers. Tibo stated that the reset is expected to go live around 2 PM Pacific Time, corresponding to approximately 5 AM Beijing Time on August 24. The team has also discovered an unrelated new optimization method, which is said to significantly improve efficiency and will continue to be implemented next week. [Original Link]