TL;DR
Get business pricing on monitors, keyboards and dev gear
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
PostgreSQL 19 is likely to arrive in November, later than the project’s usual mid-September to mid-October release window, according to developer Robert Haas. Haas argues that LLM-assisted bug reports have increased the workload for human reviewers and maintainers; the project has not confirmed that as the official cause of the delay.
PostgreSQL 19 is likely to be released in November, later than the database project’s usual mid-September to mid-October window, as maintainers handle a larger volume of bug fixes and late feature reversions, PostgreSQL developer Robert Haas wrote. Haas attributes much of the added workload to bugs reported by people using large language models (LLMs), but this is his explanation, not a stated official project finding.
Haas compared commit activity during the same period in 2025 and 2026, from April 8 through September 25. He counted 1,443 commits to the development branch in 2026, compared with 1,037 in 2025. Commits to the stable branch for the release in progress rose from 724 to 1,169. Haas said the development-branch count had declined slightly when comparing the relevant post-freeze work, while back-patched commits rose by more than 60%.
He also reported a rise in commits to a prior stable branch: during the matching period, commits to the PostgreSQL 18 stable branch numbered 433 in 2026, compared with 248 to the PostgreSQL 17 stable branch in 2025. That is a rise of more than 74% in the counts he presented. Haas said the figures reflect more bug reports from people using LLMs, while acknowledging that some of the increase could have other causes, including developers making more mistakes.
Several substantial PostgreSQL 19 features were removed late in beta, Haas said. Between August 27 and September 15, maintainers reverted UPDATE/DELETE FOR PORTION OF, SQL Property Graph Queries (SQL/PGQ), and ALTER TABLE MERGE PARTITION/SPLIT PARTITION. He described multiple late removals of major features as unusual, although he noted that reversions after the annual April 8 feature freeze are not unheard of.
More Bug Reports, More Review Work
The potential delay reflects a trade-off between a predictable release calendar and the time needed to review changes and address defects before shipping a major database release. LLMs can help identify bugs and draft fixes, Haas said, but generated fixes still require human review and may not meet PostgreSQL’s correctness or project standards. The growing volume of reports can therefore add work even when tools help uncover problems faster.
The commit counts also point to work extending beyond the upcoming release. Haas said fixes were being applied to older stable branches as well as the release under development. For users and organizations that depend on PostgreSQL, the practical impact is a later expected release and a continuing stream of maintenance work. The figures alone do not establish how many bugs were found by LLMs, how severe they were, or whether the extra commits were all prompted by AI-assisted reports.
Popular right now
Ninja CREAMi Scoop & Swirl 13-in-1 Ice Cream and Soft Serve Maker (NC701)
Shark Navigator Lift-Away Upright Vacuum, NV360
Shark PowerDetect Corded Stick Vacuum with HEPA Filtration, Purple (HZ4001AMZ)
As an affiliate, we earn on qualifying purchases.
The Release Calendar and Late Reverts
Haas said PostgreSQL major releases have typically arrived between mid-September and mid-October over the past decade. For PostgreSQL 19, he wrote that a November release was the most likely outcome and that a further postponement could not be ruled out. This is a forecast in his report; the supplied material does not include a formal release announcement setting a new date.
The project’s feature freeze took place on April 8, according to Haas. After that point, work generally focuses on stabilizing the release, though features may still be removed if maintainers judge them unsuitable to ship. Haas said that pulling major features immediately after the freeze is not especially uncommon, but that several significant reversions during the late beta period made this year unusual.
What the Commit Counts Cannot Show
The release date remains unconfirmed in the supplied reporting. Haas said November was likely and a further delay was possible, but no final schedule or official project explanation is provided. It is also unclear how much of the increase in stable-branch commits was caused specifically by LLM-assisted reports, rather than other sources of bugs or changes in reporting and maintenance activity.
The figures count commits, not unique defects, reports, or AI-generated fixes. They do not identify the severity of the underlying issues or show that every additional commit represents a separately discovered bug. Haas said he thought committers had been about as conservative as in prior years and that LLM-assisted reports were the likelier explanation, but he presented this as an informed view, not a demonstrated causal finding.
Release Timing and Review Work
PostgreSQL maintainers will need to continue reviewing fixes and deciding which changes are suitable for the release before setting or meeting a final shipping date. Based on Haas’s report, readers should expect PostgreSQL 19 in November at the earliest under his forecast, while recognizing that he did not rule out a later release. The project’s next concrete signal will be an official schedule update or release announcement.
Beyond this release, Haas said PostgreSQL developers are working out how to respond to the larger number of defects being reported and how to balance maintenance with future development. He did not outline a specific process change or timetable. Whether LLM-assisted discovery creates a sustained increase in review demands, and how the project will manage that workload, remain open questions.
Key Questions
When is PostgreSQL 19 expected to be released?
Robert Haas said a November 2026 release is most likely, with a further delay possible. The supplied material does not include a confirmed release date from the project.
Why does Haas think PostgreSQL 19 is late?
Haas attributes the delay mainly to more bug reports from people using LLMs to find defects. He says the additional reports take human time to investigate, fix, and review. That is his assessment, not a confirmed official explanation.
Which PostgreSQL 19 features were reverted?
Haas named UPDATE/DELETE FOR PORTION OF, SQL Property Graph Queries (SQL/PGQ), and ALTER TABLE MERGE PARTITION/SPLIT PARTITION. He said they were reverted between August 27 and September 15.
Do the commit figures prove that LLMs caused the increase?
No. The figures show higher commit counts in specified development and stable branches during the periods Haas compared. They do not establish the cause of each commit or count bugs found by LLMs. Haas argues that AI-assisted reports are the likeliest explanation for much of the rise.
Source: rhaas.blogspot.com
