GPT-5.6 Sol High consumed ~60% of my 5-hour Plus limit in about 6 minutes

What version of Codex CLI is running?

0.149.0

What subscription do you have?

Plus

Which model were you using?

GPT-5.6 Sol High

What platform is your computer?

Windows

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

No response

Codex doctor report

What issue are you seeing?

GPT-5.6 Sol High consumed ~60% of my 5-hour Plus limit in about 6 minutes

I’m using Codex with a ChatGPT Plus subscription.

Today I ran a single lightweight coding request using GPT-5.6 Sol with High reasoning. The task took approximately 6 minutes to complete.

Before the request, my 5-hour usage limit was close to 100%.

After this single request, approximately 60% of the entire 5-hour limit had been consumed.

This means one relatively small task used more than half of the available 5-hour quota.

Environment

• Plan: ChatGPT Plus
• Model: GPT-5.6 Sol
• Reasoning effort: High
• Number of requests: 1
• Approximate execution time: 6 minutes
• 5-hour quota consumed: ~60%
• Task complexity: lightweight / normal coding task

Why this seems abnormal

I understand that the 5-hour limit is based on compute usage and not on actual wall-clock time.

However, consuming around 60% of the entire quota in approximately 6 minutes from a single lightweight task seems far beyond reasonable usage.

At this rate, a Plus user can effectively get only around 10–20 minutes of serious work with Sol High before hitting the 5-hour limit.

This is a major regression compared with previous Codex usage and makes it very difficult to use Codex for normal software development sessions.

There are also multiple other reports from Plus users describing similar unusually fast depletion of the 5-hour limit.

Could the Codex team please confirm whether:

1. This level of usage is expected for GPT-5.6 Sol High on Plus.
2. The accounting/rate-limit system has recently changed.
3. There is currently a bug causing usage to be charged too aggressively.
4. The 5-hour quota for Plus was significantly reduced.

If useful, I can also provide screenshots of the usage meter and a Codex /feedback ID for the affected session.

What steps can reproduce the bug?

Uploaded thread: 01a04bc7-cae1-7210-8621-3e1a1e542784

What is the expected behavior?

No response

Additional information

No response

Start by reviewing the uploaded thread 01a04bc7-cae1-7210-8621-3e1a1e542784 and the affected session's /feedback ID, if available. Compare the reported usage-meter change with the request, model, plan, and reasoning effort; done means confirming whether the charge is expected or identifying an accounting defect.