Skip to content
Rush Commerce
AI & Automation3 min read

ChatGPT and Codex share one usage allowance from December

OpenAI will merge ChatGPT and Codex usage into one token-based allowance on Plus, Pro and Business in December. Your chat habits now eat your coding budget.

One pool, two products. OpenAI's help center now says that starting in December, the ChatGPT and Codex usage allowance on Plus, Pro and Business will be shared, and measured in tokens. A long research chat in the afternoon now comes out of the same budget as the Codex refactor you queued that night. Three days ago we wrote that your AI quota moved twice in 30 days. This is the third move, and it changes the unit, not only the size.

What actually happened

OpenAI updated its article Understanding ChatGPT and Codex usage limits. It ties the change to bringing the agentic features of Work into the main ChatGPT experience. The key points:

  • Plans: Plus, Pro and Business. The article does not mention Go, Enterprise or Edu.
  • Unit: tokens. How much you use depends on the model and the task. OpenAI says more advanced models and complex tasks use more.
  • Exception: Instant chat does not draw from the allowance.
  • Visibility: combined remaining usage and reset times will show in Settings → Usage.
  • At the limit: wait for the reset, upgrade, or buy credits for eligible features.
  • Expectation: OpenAI says it expects most users won't hit resets more often than they do now.

What the article does not give: a specific day in December, token amounts per plan, or new five-hour and weekly numbers. An OpenAI Community thread about the change filled up on November 1 with users who are worried their limits will drain faster. Nobody outside OpenAI knows the numbers yet, including us.

Why a shared ChatGPT and Codex allowance matters for your business

Your developers and your chat users now compete. If one Business seat does both, a heavy Codex day can block the chat work, and the reverse. Before December, find out who uses which product on each seat. Split seats where one person does both at volume.

Model choice becomes a cost decision. Token-based metering means the bigger model costs you more allowance on every task. Set a default model per task type: small model for chat drafts, big model only for hard code. Write it down, or everyone picks the biggest one.

Seat allowances are still not infrastructure. This is the same rule we keep repeating. Scheduled jobs, like nightly test runs or ticket triage, go on metered API keys with a budget cap and a fallback model. A seat can change with one help-center edit. An API invoice only changes when you change the work.

Key takeaways

  • From December, Plus, Pro and Business share one usage allowance across ChatGPT and Codex
  • Usage is measured in tokens; bigger models and harder tasks use more
  • Instant chat does not count against the allowance
  • OpenAI has not published a start day, per-plan token amounts, or new five-hour and weekly limits
  • Check who uses chat and Codex on the same seat before the change lands
  • Move scheduled agent work to metered API keys with budget caps

Know what each AI task costs before the vendor changes the meter. We build agent pipelines with per-task token logs, a default model per job, and a fallback route, so a plan change is a number on a report, not a blocked team. See how we build it or run your AI seat costs through the ROI calculator.

Sources: OpenAI Help Center, OpenAI Developer Community.

  • #openai
  • #codex
  • #chatgpt
  • #rate-limits
  • #vendor-risk
TR

Tommy Rush — Founder, Rush Commerce

Operator turned builder. 15+ years running operations — now shipping the systems businesses run on. More

Get The Rush Report weekly — one email, zero fluff.

Keep reading

AI & Automation3 min read

Sema4.ai 3.0 Verified Queries: Stop Paying to Regenerate SQL

Sema4.ai 3.0 claims Verified Queries cut LLM tokens 80% by reusing approved SQL. The pattern works in any AI agent stack. Here's how to apply it.

Read it
AI & Automation2 min read

OpenAI Cuts Funding Round to $20B at $1.25T: Plan for It

OpenAI reportedly cut its funding round from $30B to $20B at a $1.25T valuation during an AI selloff. What that means for your AI vendor costs and contracts.

Read it