GPT-5.3-Codex-Spark quota visible on Pro account but model unavailable in Codex

What version of Codex CLI is running?

codex-cli 0.154.0

What subscription do you have?

ChatGPT Pro

Which model were you using?

gpt-5.3-codex-spark

What platform is your computer?

Darwin 27.0.0 arm64 arm

What terminal emulator and version are you using (if applicable)?

VS Code integrated terminal (zsh), macOS

Codex doctor report

What issue are you seeing?

I have a ChatGPT Pro subscription. My Usage & Billing page shows dedicated GPT-5.3-Codex-Spark usage limits:

• 5-hour usage limit: 100% remaining
• Weekly usage limit: 100% remaining

However, GPT-5.3-Codex-Spark does not appear in the /model selector in Codex CLI or in the official Codex VS Code extension.

codex login status reports:

Logged in using ChatGPT

I also performed a full re-authentication using:

codex logout
codex login

and logged back in using the same ChatGPT Pro account.

When starting Codex explicitly with:

codex -m gpt-5.3-codex-spark

the CLI initially displays:

model: gpt-5.3-codex-spark medium

It also reports:

Configured service tier priority is not advertised as supported for model gpt-5.3-codex-spark and will be omitted from requests.

However, when the first inference is attempted, Codex returns:

Model metadata for gpt-5.3-codex-spark not found.
Defaulting to fallback metadata; this can degrade performance and cause issues.

followed by:

{"type":"error","status":400,"error":{"type":"invalid_request_error","message":"The 'gpt-5.3-codex-spark' model is not supported when using Codex with a ChatGPT account."}}

At the same time, /status inside Codex recognizes my separate GPT-5.3-Codex-Spark 5-hour and weekly limits, both at 100%.

Therefore, Codex recognizes the Spark usage allocation associated with my Pro account, but the model is absent from /model and server-side inference is rejected.

What steps can reproduce the bug?

Uploaded thread: 01a0a39c-0eae-7ae2-8eb2-40c96444d4df

What is the expected behavior?

No response

Additional information

No response

Start by reproducing the issue in Codex CLI 0.154.0 with a ChatGPT Pro account, checking /model, codex login status, /status, and an explicit codex -m gpt-5.3-codex-spark request. Compare the displayed entitlement and model metadata with the server rejection; done means the model is correctly available or clearly rejected consistently, with coverage for the reported account flow.