3 people reported
All time, since Sep 26.
Tiny prompt drains usage quota
“Submitting a very short, single-word prompt ("test") using the GPT-6 Luna Light model instantly consumed 6% of the available 5-hour quota.”
Is this your bug? What gets counted gets fixed. Pile on.
Paste this into your agent. It files your report, with your error and version.
I hit this bug: rapsheet.dev/sheets/codex/issues/14855/file Follow it to file my report. Show me the text before you send.
No install. No account. Public and anonymous. The link asks your agent for one report, nothing else. What it sends
Single short prompt consumes several percent of 5-hour quota. Sending a one-word message like "ok", "test" or "hi" drops the Usage remaining panel by about 2 to 7%.
Is this your bug?
File my report- Where
- Usage limits
- Versions
- 26.909.12148 to 26.924.22138
- Platforms
- macOS, Windows
- How bad
- Breaks a normal workflow.
Part of: Codex quota drains too fast19 bugs
- Usage limits deplete too fast9 people
- Usage allowance drains too fast5 people
- Quota drops to zero unexpectedly2 people
- Usage limits drain faster3 people
- 5-hour limit drains fast5 people
- Weekly usage jumps while idle6 people
- Weekly allowance drains too fast2 people
- Usage attributed to gpt-6-astra3 people
- Faster quota depletion7 people
- Usage allowance drains fast2 people
Show all 18Show only the largest 10
- Usage limits drain too fast4 people
- Usage limit drains fast5 people
- Allowances exhaust too quickly4 people
- Allowance exhausted too quickly4 people
- Reduced GPT-6 Astra allowance4 people
- Weekly quota sudden drop44 people
- Simple prompt drains usage quota5 people
- Five-hour quota drains fast5 people
Original reports4, newest first
- GPT-6 Sol Medium: single
okinteraction consumed ~7% of 5-hour Work/Codex quotaOn GitHub#49412 - Excessive quota consumption (6%) for a single-word prompt using GPT-6 Luna LightOn GitHub#49052
- Codex Desktop usage accounting mismatch: 36,318-token Luna request followed by ~6% 5-hour depletion and ~2.95M daily tokensOn GitHub#48919
- 1-3% usage used from saying hiOn GitHub#48595