DuckDB 1.5 End of Life: November 1, 2026

By Scott Bissett  ·  Published: October 9, 2026  ·  Reference guide  ·  Dates read on October 9, 2026 at DuckDB's release calendar, its community support policy and its storage and pragmas documentation (duckdb.org).

DuckDB 1.5 reaches end of life on November 1, 2026. DuckDB prints that day itself: the past releases table on its release calendar gives 1.5.0, codename Variegata, an End of life date of November 1, 2026. DuckDB 1.5.0 was released on March 9, 2026, and the most recent 1.5 patch release shipped on September 28, 2026. The long-term support line does not offer much more room: DuckDB 1.4 LTS reaches end of life on November 17, 2026. Both release lines DuckDB currently supports end within weeks of each other, so every DuckDB deployment, whether the command line shell, a Python notebook or an embedded library in a service, needs a planned move to the next release.

DuckDB's support rule, in DuckDB's words

DuckDB's release calendar explains the long-term support model in two sentences. "Starting with v1.4.0, every other DuckDB version is going to be a Long-Term Support (LTS) edition." For those versions, "the support period for community support is currently a year after the release." The same section adds that DuckLabs offers support for older LTS versions after their community support has expired. DuckDB 1.5 is not an LTS edition; its end of life date is the one on the calendar.

The community support policy on duckdb.org says which versions the DuckDB team supports at any time: the latest stable version, the latest LTS version, and the bleeding edge main branch. It also says the team is unable to guarantee that issues will be resolved within any time frame.

The current release lines

End of life dates as DuckDB prints them in its release calendar; the LTS marking is DuckDB's own. Each line links to its live status page here.

Every DuckDB release line back to 1.0 is on our DuckDB lifecycle page, with the dates re-checked at every build.

Key dates

• September 16, 2025: DuckDB 1.4.0 released, the first LTS edition.
• March 9, 2026: DuckDB 1.5.0 released.
• June 17, 2026: the most recent 1.4 LTS patch release.
• September 28, 2026: the most recent 1.5 patch release.
• November 1, 2026: end of life for DuckDB 1.5. This page.
• November 17, 2026: end of life for DuckDB 1.4 LTS.

What end of life means for DuckDB

DuckDB is an in-process database: it runs inside the program that uses it, as a Python package, a Node.js module, a JDBC driver, a library linked into a service, or the standalone command line shell. Nothing stops working on November 1, 2026. What changes is that bug fixes and security fixes are made for newer versions, and issues filed against 1.5 fall outside DuckDB's community support policy, which names the client APIs it supports first-tier: the command line shell and the C, Go, Java (JDBC), ODBC, Node.js, Python, R, Rust and WebAssembly clients.

Because DuckDB is a library, the version in use is often decided by a dependency file rather than by an administrator. A pinned version in a requirements file, a lock file or a Maven build stays on 1.5 until someone changes it, and the same release can be embedded in several services at once. Finding those pins is most of the work.

How to check what you are running

DuckDB's pragmas documentation lists two equivalent statements that return the version, and both work from any client:

PRAGMA version;
CALL pragma_version();

The first two numbers are the release line; anything starting 1.5 is on the line ending November 1, 2026, and anything starting 1.4 is on the LTS line ending November 17, 2026. For a Python environment, pip show duckdb prints the installed package version without opening a connection. Across a fleet, an SBOM lists the DuckDB package in each service; our SBOM end-of-life audit guide shows how to read one, and our EOL Checker and API resolve the result against the dates above.

Upgrade path

1. Do not move back to 1.4 LTS. Its end of life, November 17, 2026, comes only weeks after the end of 1.5, and moving back gives up features shipped in 1.5.
2. Plan for the next stable release. DuckDB's release calendar lists 2.0.0 as the next planned release and warns that the maintainers "may decide to push back release dates to ensure the stability and quality of releases." A new major version can change behaviour, so read its release notes before moving, and test the queries and client code you depend on. DuckDB's support policy also says that starting from version 2.0 it will ship a stable C++ API; until then the C++ API is described as intended for internal use.
3. Mind the database file format. DuckDB's storage documentation says that by default versions 1.0 to 1.5 create database files with storage version 64, corresponding to v1.0.0, and that newer versions aim to read files written by older ones. Files written with an explicit newer STORAGE_VERSION cannot be opened by older DuckDB releases, so keep a backup before opting into newer storage if you might need to roll back.
4. If a file will not open, export and import. The same page gives the fallback when a newer version reports an incompatible storage format: open the file with the older DuckDB and run EXPORT DATABASE 'tmp', then start the newer DuckDB and run IMPORT DATABASE 'tmp' against the new file.
5. Update every place DuckDB is pinned. Change the version in each requirements, lock or build file, rebuild, and run the tests, including any queries that load DuckDB extensions.

Frequently Asked Questions

Related

• DuckDB 1.5 — the live status page with its EOL Risk Score
• DuckDB 1.4 — the LTS line and its live status
• DuckDB — every release line's dates
• Python end of life — the runtime under most DuckDB notebooks and pipelines
• SBOM end-of-life audit — finding embedded libraries like DuckDB in an SBOM
• 2026 EOL calendar — every dated end of support this year
• DuckDB release calendar — the vendor page these dates were read from
• How we verify our dates — the rules every number on this site is held to