OpenAI doubles usage limits for ChatGPT Pro and Business Premium tiers

What changed

CryptoBriefing reported on 29 October that OpenAI is doubling usage limits for Pro 100, Pro 200, Pro 500 and Business Premium. The reversal came a day before Pro 200 allowances would have been halved from 20x to 10x the Plus baseline.Pro 200 returns to 20x immediately. Business Premium Work and Codex are also doubled. Pro 500 keeps its always-on dots agent feature on a separate allowance.

Context

The timing matters: OpenAI was a day from reducing Pro 200 allowances, effectively doubling the per-unit cost for those subscribers. Reversing suggests negative feedback or strategic recalibration. The decision follows DevDay where team workspaces, @ChatGPT for Slack/Teams and Pro 500 were launched.

For Pro 200 power users facing a 50% Work/Codex reduction, the reversal is a significant reprieve. For OpenAI, it signals competitive pressure from Claude, Gemini and others.

Caveats

Doubled limits apply to Work and Codex allowances, not API usage. Dots agents allowance is separate. The change does not affect the lowest ChatGPT Plus tier.

The report does not specify if the doubled limits are permanent or a temporary adjustment.

Tier-by-tier breakdown

The Pro 100 tier sees its Work and Codex allowances doubled, giving subscribers more room for day-to-day coding and writing tasks. The Pro 200 tier — which was facing a halving of its allowance from 20x to 10x the Plus baseline — now stays at 20x, a reversal OpenAI announced the day before the cut was scheduled to take effect.

Pro 500, the $500/month tier launched at OpenAI's DevDay in September, is also included in the doubling. This tier features always-on ChatGPT agents, called 'dots', which operate on a separate allowance unaffected by this change. Business Premium customers also benefit from doubled Work and Codex allowances.

Why this matters

OpenAI's pricing has been a sensitive topic since the introduction of multiple Pro tiers. The planned reduction in Pro 200 allowances — doubling the effective cost per unit of usage — had drawn criticism from power users who rely on the tier for heavy coding and analysis work. Reversing the cut the day before it was to take effect suggests OpenAI was responsive to that feedback or concerned about losing high-value subscribers.

The decision also reflects the competitive landscape: Anthropic's Claude, Google's Gemini and other models have been increasingly competitive on capability, and locking in Pro subscribers with generous allowances may be a retention strategy as much as a pricing decision. The company has been investing heavily in model efficiency, and the doubled limits may also signal confidence that per-token costs are coming down fast enough to make the math work.

What the tiers cost

ChatGPT Plus costs $20/month and gets basic access to OpenAI's models. Pro 100 at $100/month, Pro 200 at $200/month and Pro 500 at $500/month each step up in usage allowances and features. Business Premium is priced per seat for organizations at $200/month per user.

The doubling of limits means Pro 200 subscribers effectively get 20x the Plus baseline for Work and Codex usage — the same as before, but the key change is that OpenAI was planning to drop that to 10x on October 30. The reversal keeps the existing value proposition intact. For Pro 100 and Pro 500 subscribers, the doubling is a pure improvement to the value their subscription delivers.

The dots agents — always-on ChatGPT agents introduced at DevDay — run on their own allowance, which is unchanged. This means the new limits primarily benefit users who do heavy coding and document work within ChatGPT, rather than users of the automated agent features.

Market reaction

The announcement was covered by multiple outlets noting the unusual timing of reversing a cut the day before it was scheduled. The decision suggests OpenAI's internal data showed that the subscriber retention risk from the cut exceeded the cost savings. At $200/month, Pro 200 sits in a competitive sweet spot where users actively compare against other premium AI services.

The broader subscription market for AI has been growing rapidly, with OpenAI, Anthropic, Google, Microsoft and others all offering tiered plans. The doubling of limits at the premium tiers may trigger competitive responses from rivals, particularly Anthropic which has been positioning Claude Pro as an alternative to ChatGPT Pro for coding and analysis. The next several months will show whether usage-limiting or usage-expanding strategies dominate.

How the limits work

Each Pro tier has a monthly allowance measured in 'Work' and 'Codex' units. Work covers ChatGPT conversations, file uploads, canvas and analysis features. Codex covers the coding and agent features within ChatGPT, including the ability to run code and interact with the code interpreter. The doubling applies to both allowances proportionally.

Unused allowance does not roll over to the next month. The limits are designed for individual subscribers on each plan; Business Premium offers per-seat licensing with the same doubled limits for each seat. The always-on agents feature introduced at DevDay operates on a separate pool that is not part of the Work or Codex allowance.

Comparison to competitors

The Pro tier pricing structure is unique to OpenAI. Anthropic's Claude offers a $20/month Pro plan and a $100/month Max plan with higher usage, but does not have a $200 or $500 tier with differentiated allowances. Google Gemini Advanced runs on a $20/month Google One plan with a fixed cap.

By offering dedicated Codex and Work allowances at multiple price points, OpenAI is betting that power users — particularly developers, analysts and heavy writers — are willing to pay multiple tiers of premium subscriptions for differentiated access. The doubling of limits reinforces that bet while also acknowledging the feedback that the planned Pro 200 reduction would have hurt the most engaged subscribers.