Having this issue right now, fix it please
it always amuses me when people seem to raise question marks about using chatgpt to help draft responses about chatgpt on a chatgpt forum 
⌠like.. .why wouldnât I?
happened to me this morning once, waited 20 minutes and refreshing solved. happened to me again right now. switched to 5.4 as i cant afford to interrupt the workflow..
I also experienced the same issue with Windows Codex.
- Used up the weekly limit on the Pro account
- Changed the model to GPT-5.3-Codex-Spark
- The message appeared when logging in with the Plus account
- Rebooted the PC
- The message did not appear when logging in with the Plus account
Selected model is at capacity. Please try a different model. Still Not Resolved
I Have windows codex, 5.5 Extra High Plan Pro 5x Personal
I made an accout to report this, it is not fixed.
Selected model is at capacity. Please try a different model. Still Not Resolved
I Have windows codex, 5.5 Extra High Plan Pro 5x Personal
I have pinged the team to raise awareness that this is still ongoing for some users.
I have had same issue and i wasted so many tokens due to this issue as it was in middle of development and it just messed up all my code. So now to rectify those issues it needs to be redone. That is not acceptable due to an error from openAI 
all you have to do is wait few minutes and ask : continue.
I tried today and I lost nothing. Codex just went back to work.
Thanksfully his work is really segmented so he can start back to were he stopped.
also still an issue for me - please fix 
same issua: Reconnecting⌠4/5 (2m 27s ⢠esc to interrupt)
â Stream disconnected before completion: error sending request for url
Same issue itâs horrible can you folks please fix it I pay 200$ a month for this like loads of others. + VAT
Same here, completely breaks the flow.