Codex Desktop bug report: weekly quota dropped 67 points in 38 seconds
EgorStepkin · ghdev · 2026-10-17
Issue #54677 on openai/codex: a Windows user reports from local logs that on Oct 16 the remaining weekly Codex allowance fell from 77% to 10% between 21:10:43 and 21:11:21 Moscow time — a 67-point drop in 38 seconds — while token counters rose only 380k, 98% of it cached-input tokens.
- Weekly quota eventually hit 6%, effectively blocking development
- The author cautions this doesn't prove backend misbilling: delayed postings, usage outside inspected sessions, weighting and reconciliation must be checked against server records; raw log excerpts aren't yet attached
- Critical correlation window: UTC 18:10:43–18:11:21, plus preceding hours for delayed/batch usage postings
A concrete challenge to OpenAI's weekly quota metering worth tracking for an official response.
More from Models
- ChatGPT web can actually listen to audio, but the same model can't via API or Codex — Possible_Statement84 · 2026-10-23
- MiniMax M3.1 Flash free for 3 days on 0G, with 1M context and 512K output — MiniMax_AI · 2026-10-23
- TLA+ expert corrects Grok: TLC only exhausts finite state spaces, not all behaviors — tianyin_xu · 2026-10-23
- OpenAI blocks manual reset with 13% usage left: 'no reset needed right now' — AccomplishedSugar490 · 2026-10-23
- Meta's 816B Muse Spark Open scores 39, losing to 180B Qwen3.8-Flash-Next at 40 — BTA_Labs · 2026-10-23
- Gemini refuses to generate image in Docs, blames document size limit — viksit · 2026-10-23