TL;DR
Get business pricing on networking and server gear
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
DuckDB has released version 2.0.0 LTS, codenamed “Cinnamon Teal,” a feature-focused update built from close to 20,000 commits since version 1.5. The release adds a more stable Quack client-server protocol, asynchronous I/O, a more fully integrated VARIANT type and other features; the project says it will support 2.0 until at least October 2027.
DuckDB has released version 2.0.0 LTS, codenamed “Cinnamon Teal,” adding a more polished client-server protocol, asynchronous file I/O and expanded support for semi-structured data. The project describes it as a feature release built from close to 20,000 commits since version 1.5 and says it will support the 2.0 line until at least October 2027, giving users a long-term option as they evaluate the changes.
The release improves Quack, DuckDB’s native protocol for connecting a client to a DuckDB server, with performance and stability work, and adds a CONNECT statement for attaching to and switching to remote databases. DuckDB says CONNECT can also work with compatible systems such as PostgreSQL and MySQL. A new remote pushdown optimizer can send queries to those remote servers rather than first transferring entire tables to DuckDB. The project’s example server listens only on localhost by default; its announcement advises putting it behind a TLS-terminating reverse proxy for connections from other machines and says direct exposure through an option is intended for testing.
Asynchronous I/O is another central change. DuckDB says it can read Parquet and CSV files, as well as its own storage format, by issuing background requests and keeping multiple reads in flight. The aim is to separate I/O scaling from query processing and improve parallelism, particularly for remote data on object storage. The announcement describes the expected benefit but does not give a general benchmark figure in the supplied release material.
DuckDB 2.0 also makes its VARIANT type more fully integrated. VARIANT stores semi-structured data with differently shaped records in one column; the project says DuckDB can identify shared structure, store it in a shredded form and execute on it without treating the values as plain text JSON. The release material also lists CLI agent mode, triggers and an extensible SQL parser among the new features. It identifies these as release highlights, but the supplied details do not explain their full behavior or limitations.
A Longer Runway for New Workloads
The update matters to users who need DuckDB to work across both local and remote data. Quack and CONNECT broaden the ways a DuckDB session can reach remote databases, while query pushdown could reduce the amount of data transferred for compatible sources. The practical result will depend on the database, query and deployment, so the release announcement alone does not establish how much faster a particular workload will run.
For teams reading large files from object storage, asynchronous I/O targets a common bottleneck: waiting for remote reads to finish before more work can proceed. The more complete VARIANT implementation also gives users a columnar route for data whose fields vary from record to record. Alongside these changes, the LTS support horizon gives organizations time to test and plan an upgrade rather than treating the release as a short-lived feature branch.
DuckDB 2.0.0 LTS database management software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Three DuckDB Support Lines
DuckDB says version 2.0 is primarily a feature release; it contrasts that emphasis with DuckDB 1.0, which it characterized as focused on stability and backward compatibility. Some features were already introduced in earlier form: the project says Quack arrived in DuckDB 1.5.3 in May, and VARIANT first appeared in 1.5. Version 2.0 polishes the protocol and expands the integration of the type.
The project’s release announcement describes three supported lines. DuckDB 2.0 is planned for support until at least October 2027. Version 1.4, the previous LTS release, remains supported until late November 2026. DuckDB also extended support for version 1.5 until March 2027, effectively making it an LTS line as well. This overlap is intended to give users time to test the new release and choose a migration schedule.
Availability varies by client. The announcement says the CLI and Python releases are already available, while Go, R and Java libraries, along with some extensions, may take a few days because they require additional changes and review. DuckDB points users to its installation page, release notes and new documentation guide for details on features and migration.
““DuckDB 2.0 is above all a feature release.””
— DuckDB project, release announcement
As an affiliate, we earn on qualifying purchases.
Rollout and Migration Details
The announcement does not provide a single performance figure showing how asynchronous I/O or remote query pushdown will affect workloads across different systems. Results will depend on data sources, network conditions and query patterns. The release material also identifies agent mode, triggers and parser extensibility as features without supplying enough detail here to assess their scope, defaults or constraints.
Not every client or extension was available at announcement time. DuckDB said Go, R and Java libraries and some extensions could take a few days to ship. The precise timing for each package, and the changes users may need to make for their own applications, were not specified in the supplied announcement. Users should consult the release notes and migration guidance before upgrading, particularly because the project explicitly includes breaking changes in its 2.0 documentation.
As an affiliate, we earn on qualifying purchases.
Client Releases and Upgrade Planning
DuckDB’s next near-term step is completing the rollout of client libraries and extensions that were not ready on release day. Users can check the project’s installation page for package availability and its release page for the full change list. The new documentation guide maps the feature announcements to their technical documentation.
For teams considering an upgrade, the project’s stated support dates offer time to test version 2.0 while earlier lines remain supported. DuckDB says version 1.4 support runs until late November 2026 and version 1.5 support until March 2027; the 2.0 LTS line is planned through at least October 2027. The announcement does not set a separate date for a migration deadline, so organizations will need to assess compatibility and rollout timing against their own applications.
semi-structured data storage solutions
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What is DuckDB 2.0.0 LTS?
It is a long-term support release of DuckDB, announced October 21, 2026, and codenamed “Cinnamon Teal.” The project plans to support it until at least October 2027.
What are the main features in DuckDB 2.0?
The release highlights a more polished Quack protocol, the CONNECT statement, asynchronous I/O, a more fully integrated VARIANT type, and CLI agent mode. It also lists triggers and an extensible SQL parser.
Does DuckDB 2.0 connect to PostgreSQL and MySQL?
The project says the new CONNECT statement can point a session at compatible remote databases, including PostgreSQL and MySQL. A remote pushdown optimizer can send queries to those servers, though results depend on the source and workload.
Which DuckDB 2.0 clients are available?
At announcement time, DuckDB said the CLI and Python were available. Go, R and Java clients and some extensions could take a few more days to release.
How long will DuckDB 1.4 and 1.5 be supported?
The project says version 1.4 remains supported until late November 2026 and version 1.5 until March 2027. DuckDB 2.0 support is planned until at least October 2027.
Source: hn
Fall Picks
fall essentials
As an affiliate, we earn on qualifying purchases.
