I’m consistently running into an issue using Codex in VS Code:
Setup:
VS Code + Codex extension
Model: GPT-5.4
Mode: high
Problem:
After ~3–5 minutes of normal work, the extension starts failing:
“Selected model is at capacity” errors
Requests stop going through
Question:
Is this a known issue with the VS Code extension or model capacity handling?
Any ETA on a fix or recommended best practices to avoid this?
Welcome to the forum!
Thanks for reporting, specifically for giving an exact human readable error message, “Selected model is at capacity”
Your issue has also been reported in this Discourse topic.
You can discuss Codex-related issues here, but the official place to report them is the OpenAI Codex GitHub Issues.
In doing a quick manual search of the open issues for “model at capacity” find
The noted issue suggest using the Codex slash command /feedback but then goes on to note that it is a connectivity problem local to Australia, the issue was posted two weeks ago and has not received further updates so probably transient.
Another commonality with the other Discourse topic is the use of GPT-5.4 with High.
My recommendation would be to open a New Codex Issue as the OpenAI staff watch that list closer than this forum.
Also check OpenAI status for known system issues that the OpenAI staff have investigated and confirmed are an issue.
Until the OpenAI staff confirm it is in a issue, it will not be noted in the status page so lack of seeing an issue does not mean it is being neglected.
The specific error message “Selected model is at capacity. Please try a different model.” is from Codex, e.g.
An earlier issue from a few months ago used similar reply but gave a time to retry, that is now changed.
Which plan are you using?
https://openai.com/business/chatgpt-pricing/
https://chatgpt.com/pricing/?openaicom_referred=true
Thanks a lot for the detailed information — I really appreciate it. I’ve already gone through the GitHub issues and the status page you mentioned.
I’m currently using the Plus plan, and I’m seeing the same issue on a Pro plan as well (on another PC with a work account).
Everything was working fine yesterday, but today the problem has been happening consistently for several hours. The status page looks normal, which makes it a bit confusing.
Would appreciate any additional insight if this might be a known ongoing issue.
Just a quick update — everything is working correctly now.
Nothing changed on my side: same conversation, same context, VS Code wasn’t restarted. I had just left it running in the background while working on other tasks. After trying again later, it started working without any errors.
So it seems this might not have been an issue on my side.
This is pure speculation on my part but also based on intuition from past experience of using LLMs and being on this forum daily for a few years.
A significant change in the last 24 hours was the release of gpt-image-2
Image generation is very popular and is even one of the most popular topics on this forum. As we know people enjoy experimenting with creating images and sharing them, thus those resources that could be used for GPT-5.4 are being used for the image generation.
Again, just pure speculation; if others can provide verifiable facts related to “Selected model is at capacity” would be interested in reading them.
Yeah, I had a similar thought — that it might be a surge in usage since gpt-image-2 is pretty cool and a lot of people are probably testing it right now.
In my case, there wasn’t much I could do except wait and hope it would “fix itself” 
I also tried searching around but couldn’t find a clear or transparent thread confirming that this is an ongoing issue right now, so I decided to create one just in case others are running into the same problem.
Thanks a lot for your input, really appreciate it!
