Repository navigation
Persistent "Selected model is at capacity" error across multiple models in Codex CLI, while other accounts work #45901
Description
What version of Codex CLI is running?
codex-cli 0.154.0
What subscription do you have?
pro 5x
Which model were you using?
gpt-5.6-sol , gpt-6-astra
What platform is your computer?
No response
What terminal emulator and version are you using (if applicable)?
No response
Codex doctor report
What issue are you seeing?
Codex CLI repeatedly displays the following error:
Selected model is at capacity. Please try a different model.
Switching to another model does not resolve the issue. My other accounts
can use Codex normally, but I have not yet confirmed whether the affected
and working accounts use exactly the same model, Fast mode settings,
and other configuration.
What steps can reproduce the bug?
Uploaded thread: 01a0a95f-750e-7a10-a654-ff3fa9f3b822
What is the expected behavior?
No response
Additional information
No response
Activity
github-actions commentedon Sep 16, 2026
Potential duplicates detected. Please review them and close your issue if it is a duplicate.
Powered by Codex Action
laoguodong commentedon Sep 17, 2026
I'm experiencing a very similar persistent issue in Codex.
Starting on September 16, 2026, Codex began frequently returning:
Selected model is at capacity. Please try a different model.This now happens on almost every request when using GPT-5.6-related models, rather than as an occasional peak-capacity error. Switching between GPT-5.6 models does not reliably resolve it.
The issue appeared suddenly on September 16; Codex was working normally for me before that.
Given the persistence and the similar account-scoped reports in this thread, please investigate the affected account's Codex routing/backend capacity state rather than treating this as a normal temporary capacity spike.
lixiang4543 commentedon Sep 17, 2026
Selected model is at capacity. Please try a different model.
The reason is that after switching to the GPT6 model, this error occurred, causing all models to become unavailable. Before this switch, everything was working fine.
Yongzhang-Tan commentedon Oct 4, 2026
I am seeing the same persistent failure in a remote Codex CLI session.
Please investigate the server-side model capacity and routing behavior for this recurring CLI failure.
Yongzhang-Tan commentedon Oct 8, 2026
Update (8 Oct 2026): The problem is still reproducible after switching models in the same Codex CLI session. I tried GPT-6 Astra Ultra and GPT-6.1 Sol Max; both produced the same “Selected model is at capacity. Please try a different model.” message. Selecting “continue” repeats the failure and interrupts the conversation. This indicates the suggested model switch does not resolve the issue in this account/session. Please investigate account/session-level routing or capacity state.