Selected model is at capacity across Desktop and CLI despite 86% remaining on Pro x20 #46969

What version of the Codex App are you using (From “About Codex” dialog)?

26.915.31945

What subscription do you have?

ChatGPT Pro x20

What platform is your computer?

No response

What issue are you seeing?

Additional diagnostic update:

I have now tested Codex CLI in addition to ChatGPT Desktop Codex mode.

The exact same error occurs in Codex CLI:

"Selected model is at capacity. Please try a different model."

At this point, the issue has been reproduced across:

• ChatGPT Desktop / Codex mode
• Codex CLI
• Multiple available models
• A completely new thread
• Minimal prompts such as "Hi"

I have also encountered:

"Error running remote compact task: Selected model is at capacity. Please try a different model."

My environment/status is:

• Subscription: ChatGPT Pro x20
• Codex allowance remaining: approximately 86%
• Usage limit reached: No
• Desktop version: Powered by Codex & OWL 26.915.31945
• OS: macOS
• Workspace: Personal

AI support has already confirmed that my Codex allowance is available and that this is not a usage-limit or entitlement exhaustion issue.

Since the exact same error occurs in both the Desktop app and Codex CLI, this strongly suggests that the problem is not local to the desktop application.

Please investigate this as a potential account-level backend routing, capacity allocation, account state, or provisioning issue and escalate it to the Codex technical/infrastructure team.

This outage has now prevented me from using Codex for active development work for an extended period and has caused significant loss of working time.

Please record the full duration of this service interruption and review compensation, service credit, or another appropriate remedy for the period during which my paid Pro x20 Codex service has been unusable.

Please also let me know if there is any additional diagnostic information, request ID, logs, or account information that would help the engineering team investigate this issue.

What steps can reproduce the bug?

Additional diagnostic update:

I have now tested Codex CLI in addition to ChatGPT Desktop Codex mode.

The exact same error occurs in Codex CLI:

"Selected model is at capacity. Please try a different model."

At this point, the issue has been reproduced across:

• ChatGPT Desktop / Codex mode
• Codex CLI
• Multiple available models
• A completely new thread
• Minimal prompts such as "Hi"

I have also encountered:

"Error running remote compact task: Selected model is at capacity. Please try a different model."

My environment/status is:

• Subscription: ChatGPT Pro x20
• Codex allowance remaining: approximately 86%
• Usage limit reached: No
• Desktop version: Powered by Codex & OWL 26.915.31945
• OS: macOS
• Workspace: Personal

AI support has already confirmed that my Codex allowance is available and that this is not a usage-limit or entitlement exhaustion issue.

Since the exact same error occurs in both the Desktop app and Codex CLI, this strongly suggests that the problem is not local to the desktop application.

Please investigate this as a potential account-level backend routing, capacity allocation, account state, or provisioning issue and escalate it to the Codex technical/infrastructure team.

This outage has now prevented me from using Codex for active development work for an extended period and has caused significant loss of working time.

Please record the full duration of this service interruption and review compensation, service credit, or another appropriate remedy for the period during which my paid Pro x20 Codex service has been unusable.

Please also let me know if there is any additional diagnostic information, request ID, logs, or account information that would help the engineering team investigate this issue.

What is the expected behavior?

No response

Additional information

No response