Using a Codex proxy causes multiple Codex requests for one question
Quick fix
Upgrade cc-switch to v3.16.2 or later; the maintainer said it addresses repeated parsing failures and retries.
Symptom
Section titled “Symptom”发现会出现很多次模型调用,并且后面的很多次调用都是模型已经回答完毕之后才产生的Many model calls occur, and many later calls are generated only after the model has already finished answering.
This was reported with cc-switch 3.16.0 on macOS when using a Codex routing proxy: one question produced multiple model calls, including calls that continued after the response finished. A comment suggested a possible request loop.
The maintainer said the issue "should be resolved/implemented in v3.16.2". Related changes always include reasoning_tokens in Codex Chat usage to prevent repeated parsing failures and retries, and add truncated-stream detection. This was stated when the maintainer closed the issue; the original report did not provide post-upgrade retest results.
Upgrade cc-switch from the reported 3.16.0 to v3.16.2 or later.
After upgrading, use the same Codex question flow and check usage statistics. If calls still continue after the response ends, reopen the issue as requested by the maintainer and include the version, configuration, and reproduction steps.
Affected Versions
Section titled “Affected Versions”Source Issues
Section titled “Source Issues”This page is distilled from 1 real issues
(No FAQ yet)
Related problems
Section titled “Related problems”This is an unofficial community wiki with no affiliation to the cc-switch authors or the project itself. Its content is compiled from the project's public GitHub issues. This site distributes no software.