O/是彻底不想好好玩了，GPT被降智发邮件申诉的后续

PRO 20x ，菲区订阅，美区梯子，同办公室4人共享，三天前开始不断的overloaded、各种502、429

看了L站佬友的情况，发邮件过去申诉，说自己并没有任何违规，但是可用性大幅下降巴拉巴拉，直接得到以下回复，且目前无论web 、还是codex 都完全不可用，用就是返回：所选模型已达容量上限。请尝试其他模型。这是已经收敛到只有一个人在用的情况下了。

但明明重置了额度，目前还是100%

这个月的订阅，还有5天。。。

这个号看来是废了，后续持续报道。

以下回复邮件原文

Hello,

Thank you for contacting OpenAI Support and for sharing your experience.

We understand that you have used ChatGPT for a long time and have recently noticed what feels like a meaningful decline in response quality or reasoning capability.

We reviewed the available information associated with your request. At this time, we do not see an active integrity enforcement associated with this issue.

This means the change in quality you are experiencing is not currently documented as the result of an account warning, deactivation, or policy-related restriction.

Regarding your request to return the account to its previous “normal effort/model” behavior, OpenAI Support does not have a manual switch that can restore an individual account to an earlier model state or permanently change the reasoning behavior assigned to that account.

Check your current model and reasoning settings

If you are on an eligible paid plan, please check the model picker and confirm which model and reasoning level you are using.

GPT-5.6 Sol supports selectable reasoning levels on eligible paid plans, including:

Instant

Medium

High

Extra High

You can also check Settings > General > Higher intelligence. Where available, this setting allows ChatGPT to automatically use additional reasoning for more complex requests even when Instant is selected.

Available models and reasoning options can vary by subscription plan and workspace settings.

If the quality difference continues

If you continue seeing noticeably weaker results while using the same model and reasoning configuration, please send us one or two specific recent examples showing:

the model selected;

the reasoning level selected;

the approximate date and time;

what you asked ChatGPT to do;

what you expected from the response; and

what was materially different or worse compared with the quality you previously received.

You may remove or redact any personal, confidential, or sensitive information.

Examples using the same model and settings are especially useful because they help distinguish a genuine change in model behavior from differences caused by model selection, reasoning level, conversation context, or other product settings.

For reference, the following OpenAI resources may be helpful:

GPT-5.6 and GPT-6 Pro in ChatGPT — OpenAI Help Center

OpenAI Model Release Notes

OpenAI Status

Thank you for taking the time to report the change you have noticed. Based on the information currently available, we do not see evidence that your account has been intentionally limited because of a policy-related action.

Best regards,

Glen Anthony

OpenAI Support