GPT-5.3-Codex-Spark quota available on Pro account but model missing from Codex App selector

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

26.803.61601

What subscription do you have?

ChatGPT Pro

What platform is your computer?

Darwin 25.5.0 arm64 arm

What issue are you seeing?

Issue Description

I am a ChatGPT Pro subscriber and my account currently has a separate GPT-5.3-Codex-Spark weekly usage quota with 90% remaining.

However, GPT-5.3-Codex-Spark does not appear in the model selector in the Codex desktop app.

The models currently available in my Codex App selector are:

• GPT-5.6 Sol
• GPT-5.6 Terra
• GPT-5.6 Luna
• GPT-5.5
• GPT-5.2

GPT-5.3-Codex-Spark is completely missing.

At the same time, the Codex usage page clearly shows a dedicated “GPT-5.3-Codex-Spark usage limit” with 90% remaining and the same weekly reset date.

This appears to be an inconsistency between my account entitlement/quota and the models exposed by the Codex App.

Could you please confirm whether:

1. GPT-5.3-Codex-Spark has intentionally been removed from the Codex App model selector;
2. this is part of a staged rollout/model availability change; or
3. this is an account entitlement/model-selector synchronization bug?

<img width="972" height="442" alt="Image" src="https://github.com/user-attachments/assets/f57b3e25-936e-48a8-8406-7849810fca10" />

<img width="838" height="165" alt="Image" src="https://github.com/user-attachments/assets/9012eb5c-b014-4d5e-ac2e-c3664d372edb" />

I have attached screenshots showing both the missing model in the Codex App selector and the available GPT-5.3-Codex-Spark quota.

What steps can reproduce the bug?

Feedback ID: 019ff646-cf5f-7cd3-8c1e-9a571c8aa8ca

What is the expected behavior?

GPT-5.3-Codex-Spark should appear in the Codex App model selector for my ChatGPT Pro account while my account has an active dedicated GPT-5.3-Codex-Spark usage quota.

If Spark has intentionally been removed or temporarily disabled in the Codex App, the usage/limits page should clearly reflect that change instead of continuing to show an active separate Spark quota.

Additional information

_No response_