OpenAI temporarily removes ChatGPT Work and Codex usage limits

OpenAI has announced a temporary change to usage limits for ChatGPT Work and Codex, giving subscribers more freedom to continue longer work sessions without reaching the usual five-hour restriction.

The update applies to users on Plus, Business and Pro plans. During this temporary period, eligible subscribers will no longer be subject to the five-hour usage limit that previously applied to extended sessions. The change could benefit developers, researchers and professionals who rely on the platforms for lengthy or demanding tasks.

The announcement follows a series of recent updates from OpenAI, including the wider release of its GPT-5.6 Sol, Terra and Luna models. The latest changes appear to focus on improving access and efficiency while allowing users to make greater use of the company’s tools.

Morning. The last 48 hours of Codex and ChatGPT Work have been intense! Three important updates:

– Temporarily removing the 5 hour usage limit restriction for all Plus, Business and Pro plans
– Rolling out changes that will make GPT 5.6 Sol more efficient across the board and…

— Tibo (@thsottiaux) July 12, 2026

OpenAI has stressed that the removal of the limit is temporary. The company has not confirmed how long the change will remain in place or whether a similar policy could become permanent in the future.

Five-hour restriction temporarily removed

The temporary removal of the five-hour restriction is expected to reduce interruptions during extended ChatGPT Work and Codex sessions. Previously, users who reached the limit had to wait for their allowance to become available again, which could interrupt ongoing projects and delay time-sensitive work.

For professionals using the services throughout the working day, the change may provide greater flexibility. Developers working with Codex could continue coding, reviewing software or troubleshooting without encountering the same usage restriction. ChatGPT Work users may also be able to complete longer research, writing and productivity tasks without having to divide their work across separate usage periods.

However, OpenAI has not described the update as a permanent increase in access. Users may therefore wish to view the change as a temporary opportunity rather than a lasting adjustment to their subscription benefits. The company has also not announced when the standard restriction may return.

The temporary nature of the update means that future usage conditions could change with limited notice. OpenAI may use the period to assess demand, system capacity and user behaviour before deciding whether to introduce longer-term changes. No details have been provided about whether the company is testing alternative limits or considering different usage policies for each subscription level.

The update applies across Plus, Business and Pro plans, although the announcement does not indicate whether each plan will receive the same level of access during the temporary period. OpenAI has also not provided information about possible changes for users outside these subscription categories.

GPT-5.6 Sol becomes more efficient

Alongside the temporary removal of the five-hour limit, OpenAI is introducing efficiency improvements for GPT-5.6 Sol. The changes are intended to reduce usage when completing the same tasks, potentially allowing users to achieve more before reaching other applicable limits.

In practical terms, improved efficiency could mean that prompts, responses and longer workflows use a smaller portion of a subscriber’s available allowance. This may be particularly useful for people handling complex projects that involve repeated instructions, large amounts of information or extended conversations.

OpenAI has not yet published figures showing how much usage could be saved. As a result, it remains unclear whether users will notice a minor improvement or a more significant increase in the amount of work they can complete within their existing allowance.

Tibo, a Member of Technical Staff at OpenAI, said in a post on X that the company was still measuring the effect of the changes and would provide more information later. The statement suggests that the improvements are still being evaluated and that their impact may become clearer as more people use the updated model.

The efficiency update could complement the temporary removal of the five-hour restriction by extending the amount of work users can complete during active sessions. While the lifted limit removes one immediate barrier, reduced usage consumption may also help subscribers make better use of any remaining restrictions that continue to apply.

Further details will be needed to determine how the improvements affect different types of work. Simple questions, coding projects, document analysis and longer research tasks may not experience the same level of efficiency. OpenAI has not yet explained whether the changes apply equally across all GPT-5.6 Sol features and workloads.

Usage counters receive a fresh reset

OpenAI is also resetting usage counters following the announcement. The company said the reset would take place within an hour, allowing eligible users to begin again with a fresh usage allowance.

The reset may provide an immediate benefit to subscribers who had already used a large part of their available allowance or reached an existing limit. Instead of waiting for the normal reset period, affected users will be able to return to ChatGPT Work or Codex with their usage counters refreshed.

Combined with the temporary removal of the five-hour restriction and the efficiency improvements for GPT-5.6 Sol, the reset creates a period of expanded access for Plus, Business and Pro subscribers. Users may be able to work for longer while consuming less of their available usage than before.

The changes could make ChatGPT Work and Codex more useful for extended professional tasks, particularly when projects require several hours of continuous interaction. The update may also reduce the need for users to monitor their usage as closely during demanding work sessions.

Despite the expanded access, several details remain unclear. OpenAI has not announced an end date for the temporary removal of the limit, nor has it confirmed whether usage policies will return to their previous form once the period ends. The company has also not provided final measurements for the GPT-5.6 Sol efficiency improvements.

For now, eligible subscribers have access to a more flexible usage arrangement. The combination of a refreshed counter, fewer immediate restrictions and improved model efficiency could allow users to complete more work without the interruptions associated with the previous five-hour limit.