OpenAI doubles ChatGPT Pro and Business Premium usage limits and cancels the Pro 200 cut

3 minutes ago

6 min read

OpenAI has doubled the usage limits of its ChatGPT Pro plans and of the Business Premium tier, and it has dropped a planned reduction of the allowance attached to the $200-a-month Pro plan. The change was announced on October 29, 2026 by Thibault Sottiaux, who leads OpenAI's core products and platform work, in a post on his official X account that OpenAI framed as "Day 25" of its 28-day series of quality-of-life improvements. "We are doubling all usage limits for all Pro 100/200/500 tiers and Business Premium, effective today," Sottiaux wrote, adding that "we heard your kind feedback and won't be reducing usage limits for Pro 200, so your Work and Codex usage will stay the same."

The announcement matters most for the people who use ChatGPT as a working tool rather than as a chat assistant. The allowances that change are the ones that govern Work, OpenAI's agentic environment for longer tasks, and Codex, its coding agent, both of which consume usage far faster than ordinary conversations because a single task can run many model calls, tool invocations and file operations in sequence. For subscribers who had been hitting those ceilings during long coding sessions or multi-step agent runs, the practical effect is more headroom inside the same monthly price.

According to OpenAI's updated help-center guidance, as reported after the announcement, the doubling applies to the Work and Codex quotas of the Pro 100 and Pro 500 plans, while on Pro 200 the lower allowance that had been given to some accounts is raised to the higher level. New subscriptions already carry the updated quotas, and existing customers receive them automatically, without waiting for a renewal date. The Plus plan is unchanged. OpenAI has not published new absolute figures for the Pro tiers, so the change is expressed only as a multiple of each plan's previous allowance.

··········

WHAT OPENAI CHANGED ON EACH PLAN.

The announcement touches four paid offerings, but not in the same way. Pro 100 and Pro 500 receive a straight doubling of their Work and Codex allowances. Pro 200 is the more complicated case, because OpenAI had already split its subscribers into two groups: accounts on the original, higher allowance, and accounts that had been placed on a lower allowance ahead of a reduction that was due to reach existing subscribers at the end of October. The company's statement says Pro 200 usage "will stay the same" for Work and Codex, which means the reduction will not happen, and the reported help-center wording adds that the lower Pro 200 allowance is doubled so that those accounts match the higher one.

Business Premium, the higher seat type inside ChatGPT Business workspaces, is named in the same announcement as receiving doubled limits. OpenAI has not described the Business Premium change in more detail than that, and it has not said whether the standard Business seat is affected; the official pricing documentation continues to list the standard Business seat with the same per-model limits as Plus.

........

Plan	Monthly price	Work and Codex allowance before	After October 29, 2026
Plus	$20	Plus baseline	Unchanged
Pro 100	$100	Previous Pro 100 allowance	Doubled
Pro 200 (higher-allowance accounts)	$200	Original Pro 200 allowance, reduction scheduled	Kept, reduction cancelled
Pro 200 (lower-allowance accounts)	$200	Reduced Pro 200 allowance	Doubled to the higher level
Pro 500	$500	Previous Pro 500 allowance	Doubled
Business Premium	Per seat	Previous Premium allowance	Doubled

........

The prices in the table are the monthly list prices shown in OpenAI's official ChatGPT pricing documentation, which lists Pro "from $100/month" with plans at $100, $200 or $500 and Plus at $20 per month. None of the plan prices changed with this announcement.

··········

THE PRO 200 REDUCTION THAT WILL NOT HAPPEN.

The cancelled cut explains why the announcement drew so much attention among heavy Codex users. Since late September, Pro 200 subscribers had been discussing on OpenAI's developer forum an email in which, according to their account, OpenAI said that new Pro 200 subscribers would receive a lower Work and Codex allowance and that existing subscribers would keep the higher one only through October 29, with the reduction starting on October 30. One subscriber described the change as a move from 20 times to 10 times the Plus allowance. OpenAI's October 29 post did not restate those multiples, so they should be read as subscribers' description of the earlier notice rather than as figures OpenAI has confirmed publicly.

What OpenAI has confirmed is the outcome. Existing Pro 200 subscribers who were about to lose part of their allowance keep it, and accounts that had already been placed on the lower level are lifted to the higher one. In effect, the $200 plan returns to a single allowance for everyone on it, and that allowance is the higher of the two that had coexisted since September. The timing is not incidental: the announcement arrived on the last day of the protection period that subscribers had described, one day before the reduction would have applied.

··········

WHAT A DOUBLED QUOTA MEANS IN PRACTICE.

OpenAI says that doubling a quota does not double the amount of work a subscriber can complete. Consumption in Work and Codex depends on the model selected, the type of task and the settings used, so the same allowance can cover very different numbers of tasks depending on how it is spent. The official Codex pricing documentation illustrates the spread with the estimated number of local Codex messages a Plus subscriber can send in a five-hour window, which also serves as the reference point for the Pro multiples.

........

Model	Plus, estimated local messages per 5 hours
GPT-6 Astra	5-45
GPT-6.1 Sol	15-160
GPT-6.1 Terra	35-285
GPT-6 Sol	15-150
GPT-6 Luna	350-3,000

........

The ranges are OpenAI's own estimates and are wide because a short edit and a long refactoring task consume very different amounts of compute. The same documentation states that Pro plans currently have no five-hour limit, although weekly limits may still apply, and that cloud tasks may use more of the allowance than local messages. Speed settings change the arithmetic as well: OpenAI says Fast mode uses 2.5 times the included usage of Standard mode, and Ultrafast, available on Pro 500 and eligible Enterprise and Edu plans, uses 8 times the Standard rate for GPT-6 Astra and GPT-6.1 Sol. A Pro 500 subscriber who runs everything in Ultrafast will therefore still burn through a doubled allowance much faster than one working in Standard mode.

Model choice is the other lever. The pricing table shows that the gap between GPT-6 Astra and GPT-6 Luna on the Plus plan is about 67-fold at the top of each range, 45 messages against 3,000, and OpenAI has presented its recent GPT-6.1 models, covered in our report on GPT-6.1 Terra and its API pricing, as more efficient than their predecessors. Sottiaux's post points in the same direction, saying OpenAI will "keep shipping better models and other efficiency improvements that increase your effective usage over time."

··········

WHAT CHANGES FOR CHATGPT AND CODEX USERS.

For a subscriber already on Pro 100 or Pro 500, nothing needs to be done: the larger allowance applies to the existing subscription, and new subscribers receive it from the start. A Pro 200 subscriber who joined after the September change and had been working with the lower allowance should now see the same allowance as long-standing Pro 200 accounts, and a long-standing subscriber no longer faces a reduction. Plus subscribers get no change in Work or Codex limits, which widens the gap between the $20 plan and the cheapest Pro tier.

The announcement also covers dots, OpenAI's always-on agents. Sottiaux said OpenAI will "continue offering extended limits for dots for now" and that dot usage "will still continue to be separate from your default usage." The phrase "for now" leaves the duration open, so subscribers who rely heavily on dots should not treat the extended limits as a permanent entitlement.

The change is separate from the API credits OpenAI introduced a day earlier for ChatGPT Pro and Business subscribers, described in our article on the monthly API credits of up to $500. Those credits are meant for API usage, while the October 29 change raises the in-product allowances and does nothing for API billing. A developer who uses both Codex inside ChatGPT and the API for automation now gets more capacity on both sides, but through two different mechanisms with different rules.

··········

WHAT IS ESTABLISHED AND WHAT REMAINS OPEN.

The established facts are limited but clear. OpenAI, through its head of core products and platforms, announced on October 29, 2026 that Work and Codex usage limits are doubled for Pro 100, Pro 200 and Pro 500 and for Business Premium, effective the same day; the planned reduction for Pro 200 is cancelled; Plus is unchanged; extended dot limits continue for the time being; and plan prices remain at $20 for Plus and $100, $200 and $500 for the three Pro tiers.

Several points remain undocumented. OpenAI has not published absolute allowances for the Pro tiers or for Business Premium, so subscribers cannot verify the doubling against a published number and will have to judge it from their own usage. The company has not said whether the higher limits are permanent or how long the extended dot limits will last. It has also not explained how the Business Premium change interacts with workspace-level controls set by administrators. Until OpenAI publishes concrete figures, the most reliable measure of the change is the one each subscriber can observe directly: how much longer a typical Codex or Work session runs before reaching the plan's ceiling.

··········

·····

DATA STUDIOS

·····

datastudios.org