# Release notes > Latest updates, new features, and improvements to MotherDuck. Welcome to our release notes, we're excited to hear about your experience ๐Ÿ˜ƒ :::info ๐Ÿ’ If you have any questions, please connect with us directly in our [Community Slack support channel](https://slack.motherduck.com/) or send a note to support@motherduck.com. ::: For older updates, see the [release notes archive](/about-motherduck/release-notes-archive/). ## May 22, 2026 - **Postgres wire protocol endpoint support for Looker:** The [Postgres Endpoint](https://motherduck.com/docs/getting-started/interfaces/client-apis/other/postgres-endpoint-jdbc/), a Postgres wire protocol-compatible MotherDuck client, now enables connectivity with [Looker](https://motherduck.com/docs/integrations/bi-tools/looker/). - **DuckDB & MotherDuck plugin for Obsidian:** Run DuckDB SQL or MotherDuck queries from inside an [Obsidian](/integrations/dev-tools/obsidian/) note, with results frozen as a markdown table directly below the query. Local DuckDB blocks run in Wasm with no account required; `motherduck` blocks reach your MotherDuck account for [hybrid execution](/key-tasks/running-hybrid-queries/) against shared databases. Install from the [Obsidian Community Plugins](https://community.obsidian.md/plugins/duckdb-motherduck) directory. - **Manual member deprovisioning:** Organization admins can now [deprovision and reactivate members](/key-tasks/managing-organizations/#deprovisioning-users) directly from the Members page in Settings in the MotherDuck UI, without needing [SCIM](/key-tasks/authenticating-and-connecting-to-motherduck/authenticating-to-motherduck/scim/) or an external identity provider. - **Compound file extensions in Add Data:** The [Add Data](/key-tasks/loading-data-into-motherduck/loading-data-from-cloud-or-https/) UI now recognizes compound suffixes such as `.parquet.zst`, `.csv.zst`, and `.parquet.snappy` when detecting file types from object storage. - **TOON-encoded responses from the MCP server:** Responses from the [MotherDuck remote MCP server](/sql-reference/mcp/) are now encoded in [TOON](https://github.com/toon-format/toon), a compact, token-efficient format for LLM consumption, reducing payload size on large result sets. - **Pin a Dive version when embedding:** [Embed sessions](/key-tasks/ai-and-motherduck/dives/embedding-dives/) now accept an optional `version` parameter so production embeds keep rendering a vetted Dive version even when the source Dive is edited. Omit it to follow the current version. - **Dive Viewer in ChatGPT:** The [MCP Dive Viewer](/sql-reference/mcp/view-dive/) is now available as a [ChatGPT app](https://chatgpt.com/apps/motherduck/asdk_app_696a54f1c91c81919002b9153ce0e336), rendering Dives inline in conversations. - **"Open in MotherDuck" preserves Dive state:** Clicking "Open in MotherDuck" from the [MCP Dive Viewer](/sql-reference/mcp/view-dive/) now carries your current filters and selections through to the full MotherDuck UI. - **Dive Viewer auto-refreshes stale tokens:** [Dives](/key-tasks/ai-and-motherduck/dives/) left open in browser tabs for hours no longer show "Invalid Token" errors. The session token now refreshes silently when the tab regains focus. - **Support for DuckDB client versions 1.1.x and 1.2.x has ended.** Refer to the [version lifecycle schedule](/troubleshooting/version-lifecycle-schedules/) for supported versions. ## May 15, 2026 - **SCIM provisioning:** MotherDuck now supports SCIM, so you can manage MotherDuck users from your identity provider. Available on **Business** and **Enterprise** plans. Refer to the [documentation](/key-tasks/authenticating-and-connecting-to-motherduck/authenticating-to-motherduck/scim/) for more details. - **Per-session database override for Dives:** [Embedded Dives](/key-tasks/ai-and-motherduck/dives/embedding-dives/#override-required-databases) and the [`view_dive` MCP tool](/sql-reference/mcp/view-dive/) now accept a `required_resources` parameter that overrides the Dive's source-declared `REQUIRED_DATABASES` for the session. This lets you point a single Dive at different databases per tenant or environment. - **Share Dives with their current view:** Share a [Dive](/key-tasks/ai-and-motherduck/dives/#sharing-the-current-view) along with its interactive state โ€” filters, drill-downs, sort orders, selected views. In the MotherDuck UI, the state is captured in the URL fragment, so sharing the URL lands recipients on the same view. In [embedded Dives](/key-tasks/ai-and-motherduck/dives/embedding-dives/#handle-dive-state-updates-from-embedded-dives), state changes emit `postMessage` events your host page can listen for and persist in your application. - **Preconfigure Dives with a starting UI state:** A new `initial_state` parameter on [embed sessions](/key-tasks/ai-and-motherduck/dives/embedding-dives/#preconfigure-the-starting-ui-state) and the [`view_dive` MCP tool](/sql-reference/mcp/view-dive/) loads the Dive with specific filters and selections already applied. - **Data exports from Dives in MCP apps:** The [MCP Dive Viewer](/sql-reference/mcp/view-dive/#data-exports-from-the-dive-viewer) now lets users export Dive results as CSV, Parquet, or XLSX. - **Claude Cowork support for the Dive Viewer:** The [MCP Dive Viewer](/sql-reference/mcp/view-dive/) now renders Dives inline in Claude Cowork, alongside existing MCP client support. ## May 7, 2026 - **Data exports from Dives:** [Dives](/key-tasks/ai-and-motherduck/dives/) can now include export buttons that deliver CSV, JSON, Parquet, or Excel files, both in the MotherDuck UI and in [embedded Dives](/key-tasks/ai-and-motherduck/dives/embedding-dives/#handle-data-exports-from-embedded-dives). Wire up exports from your Dive code with the new [`exportAs` and `useExport` hooks](/sql-reference/motherduck-sql-reference/ai-functions/dives/use-sql-query/#export-query-results) from `@motherduck/react-sql-query`. Embedded exports are delivered to your host page through a `postMessage` channel and require dual mode, now the default for embedded Dives. - **MotherDuck extends cloud coverage to Oregon `us-west-2`:** MotherDuck is now [available on AWS in Oregon `us-west-2`](/concepts/architecture-and-capabilities/#the-motherduck-cloud-service); users are able to create new Organizations in Oregon for lower latency and regional data residency. - **Drizzle support via the Postgres endpoint:** You can now connect to MotherDuck from [Drizzle](https://orm.drizzle.team/) using the [Postgres endpoint](https://motherduck.com/docs/key-tasks/authenticating-and-connecting-to-motherduck/postgres-endpoint/drizzle/). ## May 1, 2026 - **Streamlined Dives sidebar:** The [Dives](/key-tasks/ai-and-motherduck/dives/) list in the Object Explorer is now capped at your 10 most recent Dives, with a new **View all Dives** link that opens the full searchable Dives table. - **Wasm Client SDK no longer requires COI headers:** The [MotherDuck Wasm client](/sql-reference/wasm-client/) SDK (now at version `1.5.2`) no longer requires Cross-Origin Isolation (COI) headers, so custom DuckDB Wasm applications using `LOAD motherduck` can run in standard, non-COI environments. This change also brings faster transfers for larger payloads. - **JIT provisioning enabled by default for SSO:** When you activate [SSO](/key-tasks/authenticating-and-connecting-to-motherduck/authenticating-to-motherduck/sso-setup/) for the first time, [Just-in-Time (JIT) provisioning](/key-tasks/authenticating-and-connecting-to-motherduck/authenticating-to-motherduck/sso-setup/#just-in-time-jit-user-provisioning) is now on by default, so users from your verified domain can join the organization automatically on first login. - **Link handling in Dives:** When you click a link inside a [Dive](/key-tasks/ai-and-motherduck/dives/), MotherDuck surfaces a confirmation popup for external URLs so you can review the destination before opening it, and navigates directly for links that point inside the MotherDuck UI (for example, another Dive). [Embedded Dives](/key-tasks/ai-and-motherduck/dives/embedding-dives/#handle-link-navigation-from-embedded-dives) forward each click to the host page through a `postMessage` channel, so the parent app can apply its own navigation policy. ## April 22, 2026 - **Duckling Overview UI:** MotherDuck Admins now have access to a [Duckling Overview page](/getting-started/interfaces/motherduck-quick-tour/#duckling-overview) in Settings for a view of activity across every [Duckling](/about-motherduck/billing/duckling-sizes/) in the organization over the last 24 hours, including key metrics like status, disk spills, active minutes, and query-level drilldowns. - **Dive Viewer Inline Preview for Remote MCP Server:** The [MotherDuck remote MCP server](/sql-reference/mcp/) now includes an inline preview [Dive Viewer](https://motherduck.com/docs/key-tasks/ai-and-motherduck/dives/?_gl=1*10naj7t*_up*MQ..*_ga*MTE5NTMzMTIwMS4xNzc2ODk2MDQ0*_ga_L80NDGFJTP*czE3NzY4OTYwNDMkbzEkZzAkdDE3NzY4OTYwNDMkajYwJGwwJGgxMTY1MDY4NDcx#inline-preview-with-the-dive-viewer), allowing users and AI agents to create and view [Dives](/key-tasks/ai-and-motherduck/dives/) directly in their chat workflow. - **Removed Cross-Origin Isolation (COI) requirement for DuckDB Wasm:** MotherDuck's DuckDB Wasm integration can now run in standard i-frames and non-COI third-party environments. This is already powering the MotherDuck UI and [Embedded Dives](/key-tasks/ai-and-motherduck/dives/embedding-dives/), which default to [dual execution](/key-tasks/running-hybrid-queries/) for zero-latency client-side queries. Support for `LOAD motherduck` from custom DuckDB Wasm applications will follow in a future WASM SDK release. ## April 16, 2026 - **DuckDB 1.5.2 support:** MotherDuck supports DuckDB 1.5.2, a bugfix release. Learn more in the [official DuckDB Labs 1.5.2 announcement](https://duckdb.org/2026/04/13/announcing-duckdb-152) and [changelog](https://github.com/duckdb/duckdb/releases/tag/v1.5.2). - **DuckLake 1.0 support:** MotherDuck supports DuckLake 1.0. Learn more in the [official DuckDB Labs DuckLake 1.0 announcement](https://ducklake.select/2026/04/13/ducklake-10/). - **Concurrent checkpoints:** Checkpoints can now run concurrently to reads, insertions and deletions. Previously they could block or be blocked by user queries and also interfere with share updates, as these require a checkpoint. Note that concurrent checkpoints require writes or deletions to be issued by DuckDB clients that are at least on [version 1.5](https://motherduck.com/docs/troubleshooting/version-lifecycle-schedules/). ## April 9, 2026 - **Postgres wire protocol endpoint support for PowerBI:** The [Postgres Endpoint](https://motherduck.com/docs/getting-started/interfaces/client-apis/other/postgres-endpoint-jdbc/), a Postgres wire protocol-compatible MotherDuck client, now enables connectivity with [PowerBI](https://motherduck.com/docs/integrations/bi-tools/powerbi/). ## April 3, 2026 - **[Preview] Embedded Dives:** MotherDuck Dives can now be embedded directly in your customer-facing applications. Create a Dive, generate an embed session, and drop an iframe into your app. Your end-users can explore data in real time, powered by MotherDuck. Embedded Dives currently require a Business plan. Read the [announcement blog](https://motherduck.com/blog/introducing-embedded-dives/) for more details or refer to the [documentation](https://motherduck.com/docs/key-tasks/ai-and-motherduck/dives/embedding-dives/) to get started. - **Airbyte Connector Certified:** MotherDuck is now a certified Airbyte destination. Refer to the [Airbyte destination documentation](https://docs.airbyte.com/integrations/destinations/motherduck) for more details. - **Customizable Sidebar:** The Object Explorer sidebar is now fully customizable. Right-click any section to show, hide, or reorder it โ€” or access the customization dialog from the command menu or the org menu. Your expand/collapse state is persisted across sessions. ## March 25, 2026 - **Configurable Duckling cooldown periods:** The cooldown period for Standard, Jumbo, Mega, and Giga Ducklings can now be configured from 1 minute to 24 hours via the UI, SQL, or REST API. Learn more in the [Duckling sizes](https://motherduck.com/docs/about-motherduck/billing/duckling-sizes/) documentation. - **SHUTDOWN and SHUTDOWN TERMINATE commands:** Two new SQL commands give you direct control over Duckling lifecycle without waiting for the cooldown period. [SHUTDOWN](https://motherduck.com/docs/sql-reference/motherduck-sql-reference/shutdown-terminate/#shutdown) shuts down a Duckling after running queries complete; [SHUTDOWN TERMINATE](https://motherduck.com/docs/sql-reference/motherduck-sql-reference/shutdown-terminate/#shutdown) force-terminates immediately. Both are subject to a 1-minute billing minimum and are useful for cost control in batch pipelines or CI/CD workflows where you want to stop billing as soon as work is done. See the [command reference](https://motherduck.com/docs/sql-reference/motherduck-sql-reference/shutdown-terminate/) for details. - **Dives Version History in UI:** Every time you update a [Dive](https://motherduck.com/docs/key-tasks/ai-and-motherduck/dives/), MotherDuck saves a version. Browse previous versions using the version picker in the top-right corner of any Dive or retrieve them programmatically with [list_dives](https://motherduck.com/docs/sql-reference/mcp/list-dives/) and [read_dive](https://motherduck.com/docs/sql-reference/mcp/read-dive/). Version browsing is read-only โ€” selecting an older version does not overwrite the latest. ## March 24, 2026 - **DuckDB 1.5.1 support:** MotherDuck supports [DuckDB 1.5.1](https://duckdb.org/2026/03/23/announcing-duckdb-151) ("Variegata"). This release brings significant performance improvements including bloom filter join pushdown, stats-only min/max evaluation, faster TopN queries with late materialization, and lazy view binding. Note that the VARIANT data type and background checkpointing improvements are not yet supported in MotherDuck and will be made available in a future update. See the [version lifecycle schedules](/troubleshooting/version-lifecycle-schedules/) for supported version ranges. - **DuckLake 0.4:** The DuckLake open table format has been updated to [version 0.4](/integrations/file-formats/ducklake/#whats-new-in-ducklake-10), introducing deletion inlining, sorted compaction, stats-only `COUNT(*)`, TopN file pruning, expression-based default values, and macro support. ## March 12, 2026 - **Postgres wire protocol endpoint support for Tableau Cloud:** The [Postgres Endpoint](https://motherduck.com/docs/getting-started/interfaces/client-apis/other/postgres-endpoint-jdbc/), a Postgres wire protocol-compatible MotherDuck client, now enables connectivity with [Tableau Cloud](https://motherduck.com/docs/integrations/bi-tools/tableau/tableau-cloud/). - **Dives can now be remixed:** A new 'Remix this Dive' menu option in the left-side object explorer opens ChatGPT or Claude with pre-filled metadata to help you explore and iterate on existing analysis and create new [Dives](/key-tasks/ai-and-motherduck/dives/). ![Image](useBaseUrl('/img/about-motherduck/remix-dive.png')) ## March 6, 2026 - **Postgres wire protocol endpoint:** Query your MotherDuck databases using any Postgres-compatible client, including `psql`, Python libraries like psycopg2/psycopg3, JDBC drivers, and serverless platforms like Cloudflare Workers โ€” without installing a DuckDB client library. The Postgres endpoint is ideal for serverless environments, languages without a DuckDB SDK, thin client architectures, or any tool that supports PostgreSQL data sources. See guides for [Python](https://motherduck.com/docs/getting-started/interfaces/client-apis/python/postgres-endpoint/), [Java (JDBC)](https://motherduck.com/docs/getting-started/interfaces/client-apis/other/postgres-endpoint-jdbc/), and [Cloudflare Workers](https://motherduck.com/docs/getting-started/interfaces/serverless/cloudflare-workers/). ## February 27, 2026 - **Write support for the MCP server is now live:** Users can now perform write operations via the MCP Server using the new `query_rw` tool. This enables programmatic data modification workflows in addition to read queries. Refer to the [documentation](/sql-reference/mcp/query-rw/) and [setup guide for read-only access](/sql-reference/mcp/#restricting-to-read-only-access) for more details. ## February 26, 2026 - **Dives are now available on all plans:** Dives are now available on all MotherDuck plans, giving users access to shareable visualizations built by AI agents and backed by composable SQL. Refer to the [documentation](/key-tasks/ai-and-motherduck/dives/) for more details. - **Faster edits for Dive Previews in Claude:** Claude agents in [Claude Web](http://claude.ai/) and [Claude Desktop](https://code.claude.com/docs/en/desktop) now apply edits to the existing Dive preview instead of re-generating it from scratch. - **SSO support:** MotherDuck now supports Single Sign-On (SSO) with Okta, Microsoft Entra ID, and federated SAML/OIDC. Refer to the [documentation](/key-tasks/authenticating-and-connecting-to-motherduck/authenticating-to-motherduck/sso-setup/) for more details. ## February 19, 2026 - **Visualizations with Dives:** Create interactive visualizations directly in your MotherDuck UI using natural language. To get started, connect your [AI agent](/getting-started/mcp-getting-started/) (Claude, ChatGPT, Cursor, or any MCP-compatible client) to the [MotherDuck MCP Server](/sql-reference/mcp/) and ask it to build a Dive. The agent writes the SQL, configures the charts, and saves the visualization in your MotherDuck UI. Dives stay live and up-to-date and can be shared across your organization. Read the [announcement blog](https://motherduck.com/blog/duck-dive-and-answer/) and get started in the [Dives documentation](/key-tasks/ai-and-motherduck/dives/). Dives are now available on all MotherDuck plans at no additional charge. ## February 5, 2026 - **Expanded DuckDB Extension support in the MotherDuck MCP Server:** The MotherDuck Remote MCP Server is now compatible with the DuckLake, Spatial, Iceberg, and Delta DuckDB extensions. Read more about using the MotherDuck MCP Server [in the documentation](/key-tasks/ai-and-motherduck/mcp-workflows/). - **Point-in-time Restore**: Restore databases to previous states using automatic or named snapshots within a configurable retention window of up to 90 days. Use [`UNDROP DATABASE`](/sql-reference/motherduck-sql-reference/undrop-database/) to recover deleted databases. Learn more in the [launch blog post](https://motherduck.com/blog/point-in-time-restore/) and the [data recovery documentation](/concepts/data-recovery/). - **DuckDB Database File Upload:** Upload DuckDB database files (.duckdb, .db) from your laptop to MotherDuck using the "Add data" menu in the MotherDuck UI. Preview tables and schemas in the UI before copying to MotherDuck. See the [documentation on loading DuckDB databases into MotherDuck](/key-tasks/loading-data-into-motherduck/loading-duckdb-database/) to learn more about using DuckDB database files. ## January 29, 2026 - **DuckDB 1.4.4:** MotherDuck supports DuckDB 1.4.4, a bugfix release. Learn more in the [official DuckDB Labs 1.4.4 announcement](https://duckdb.org/2026/01/26/announcing-duckdb-144.html) and [changelog](https://github.com/duckdb/duckdb/releases/tag/v1.4.4). ## January 23, 2026 - **Expanded MCP Server support:** The MotherDuck remote MCP Server now supports [Warp](https://www.warp.dev/), [PearAI](https://trypear.ai/), [Trae](https://www.trae.ai/), [Void](https://voideditor.com/), [Positron](https://positron.posit.co/), [Supermaven](https://supermaven.com/), [Aider](https://aider.chat/), and [JetBrains IDEs](https://www.jetbrains.com/). Use your favorite AI assistant to answer questions about your data through natural conversation. See the [MCP Server documentation](/sql-reference/mcp/) to get started. - **Add Data from Cloud Storage:** Import data from Amazon S3, Google Cloud Storage, Cloudflare R2, and others directly in the MotherDuck UI. Click "Add data" and select "From cloud storage" to browse your bucket, select files (or use Wildcard mode for patterns), preview the data, and create tables. Learn more in the [documentation for loading data from Cloud Storage](/key-tasks/loading-data-into-motherduck/loading-data-from-cloud-or-https/). ## January 8, 2026 - **Giga Ducklings on Business plan:** Users on any MotherDuck Business plan can now access [Giga Ducklings](../billing/duckling-sizes/#giga), our largest compute Duckling size, built to tackle the largest, toughest, most complex data transformations. Configure your Duckling size in [Settings > Ducklings](https://app.motherduck.com/settings/ducklings). ## December 17, 2025 - **MotherDuck MCP Server:** Your favorite AI assistant can now talk directly to your data. Connect Claude, ChatGPT, Cursor, or any MCP-compatible client to MotherDuck using the MotherDuck **remote** MCP Server at `https://api.motherduck.com/mcp`. Your agent can explore schemas, run read-only SQL queries, and answer questions about your databases through natural conversation. Learn more in the [announcement blog](https://motherduck.com/blog/analytics-agents), and [MCP Server documentation](/sql-reference/mcp/). ## December 16, 2025 - **DuckDB 1.4.3:** MotherDuck supports DuckDB 1.4.3, a bugfix release. Learn more in the [official DuckDB Labs 1.4.3 announcement](https://duckdb.org/2025/12/09/announcing-duckdb-143) and [changelog](https://github.com/duckdb/duckdb/releases/tag/v1.4.3). - **PlanetScale Postgres integration:** Users of PlanetScale Postgres can now use [pg_duckdb](/concepts/pgduckdb/) to push analytical queries to MotherDuck. Analytical queries are accelerated up to 200x faster with MotherDuck, and keep your Postgres cluster optimized for transactions. Learn more in the [announcement blog](https://motherduck.com/blog/motherduck-planetscale-integration), and [integration documentation](/integrations/databases/planetscale). - **MotherDuck destination for Artie CDC**: Artie now supports MotherDuck as a destination for CDC. Users of Artie can now stream changes from OLTP databases like PostgreSQL, MySQL, and MongoDB to MotherDuck in real-time. Learn more in the [announcement blog](https://motherduck.com/blog/motherduck-artie-integration/), and [Artie documentation](https://www.artie.com/docs/destinations/motherduck). - **Recent Queries added to `MD_INFORMATION_SCHEMA`:** Organization admins on MotherDuck Business plans can now access a more realtime view of all currently running or recently completed queries across their full organization using the [`RECENT_QUERIES` view](/sql-reference/motherduck-sql-reference/md_information_schema/recent_queries/). This view offers detail for queries not yet captured in the [`QUERY_HISTORY` view](/sql-reference/motherduck-sql-reference/md_information_schema/query_history/). Both views are accessible in the [`MD_INFORMATION_SCHEMA`](/sql-reference/motherduck-sql-reference/md_information_schema/introduction/). - **New columns for query attribution in query history:** The [`QUERY_HISTORY` view](/sql-reference/motherduck-sql-reference/md_information_schema/query_history/) along with the new [`RECENT_QUERIES` view](/sql-reference/motherduck-sql-reference/md_information_schema/recent_queries/) in the [`MD_INFORMATION_SCHEMA`](/sql-reference/motherduck-sql-reference/md_information_schema/introduction/) now contain `session_name` and `duckling_id` columns, making it easy to identify which Duckling executed each query, and group read scaling queries by [`session_name`](/key-tasks/authenticating-and-connecting-to-motherduck/read-scaling/#session-affinity-with-session-name). - **MotherDuck Wasm SDK 0.8:** The [MotherDuck Wasm Client](https://www.npmjs.com/package/@motherduck/wasm-client) now leverages a different mechanism for loading the MotherDuck Wasm extension, which makes it easier to control which version of the extension is loaded. Refer to the [documentation](/sql-reference/wasm-client/) to learn more. ## December 12, 2025 - **Query Scheduling Improvements:** Small queries now complete faster without getting stuck waiting behind large, resource-intensive queries, even when heavy queries are processing in the background - **Search Enhancements:** The search bar in the top left pane of the Object Explorer can now be used to search for schemas, tables, and columns in addition to databases and [shares](/key-tasks/sharing-data/) - **Dvorak Keyboard Support:** Dvorak keyboard shortcuts are now supported in the MotherDuck UI - **Column Comments added to the Table Summary:** In the Table Summary, users can now hover over any column name to view its comments alongside the column name and type ![Image](useBaseUrl('/img/about-motherduck/column-comments.png')) ## December 4, 2025 - **Transient storage filter in Settings:** The Databases page in Settings in the MotherDuck UI now supports filtering by [storage type](/concepts/storage-lifecycle/#transient-databases) - **`DESCRIBE` and `SUMMARIZE` exports:** Downloading the results of `DESCRIBE` and `SUMMARIZE` queries is now supported in the MotherDuck UI - **DuckLake option in the Add Database menu:** MotherDuck users can create a new [DuckLake](/integrations/file-formats/ducklake/) in the 'Add Database' modal in the left hand pane of the object explorer in the MotherDuck UI - **Inline Docs are now available in the Query Editor:** Notebook cells in the MotherDuck [query editor](/getting-started/interfaces/motherduck-quick-tour/#inline-docs) provide function information on hover, showing function signatures, parameter types, return types, and descriptions without leaving the notebook. Inline Docs can be toggled on and off by going to the Preferences page in Settings. ![Image](useBaseUrl('/img/about-motherduck/ui-inline-docs.png')) ## November 14, 2025 - **DuckDB 1.4.2:** MotherDuck supports DuckDB 1.4.2, a bugfix release. Learn more in the [official DuckDB Labs 1.4.2 announcement](https://duckdb.org/2025/11/12/announcing-duckdb-142) and [changelog](https://github.com/duckdb/duckdb/releases/tag/v1.4.2). - **Full command menu now at `Cmd/Ctrl+K`:** Access common MotherDuck UI actions from your keyboard, including generating query edits, adding notebook cells, creating notebooks, and navigating between pages. Open the command menu with `Cmd/Ctrl+K` and search for options. For quick access to [generate query edits](../../key-tasks/ai-and-motherduck/ai-features-in-ui#automatically-edit-sql-queries-in-the-motherduck-ui), use `Cmd/Ctrl+Shift+E`. (Note: `Cmd/Ctrl+Shift+P` no longer opens the command menu.) - **Run queries across multiple notebooks:** You can now run cells across multiple MotherDuck UI notebooks, and allowing each to queue and run. Hover over any notebook in the left sidebar to see how many cells are running or queued. Query cancellation is also more reliable across all notebooks. ![Image](useBaseUrl('/img/about-motherduck/multiple_notebook_run.png')) ## November 6, 2025 - **MotherDuck extends cloud coverage to Europe:** MotherDuck is now [available on AWS in Frankfurt `eu-central-1`](/concepts/architecture-and-capabilities/#the-motherduck-cloud-service); users are able to create new Organizations in Europe for lower latency and regional data residency - **Expanded AI functions support for `PROMPT()`:** The [prompt](/sql-reference/motherduck-sql-reference/ai-functions/prompt/) function now supports additional parameters; MotherDuck users can now interact with Large Language Models (LLMs) directly from SQL with more customization and improved support for struct arrays, timestamps, and date and time values - - **`return_type`:** Generate strongly-typed outputs by specifying the exact SQL type to return - **`reasoning_effort`:** Use GPT-5 models with the prompt() function - **MotherDuck Wasm SDK 0.7.0:** The [MotherDuck Wasm Client](https://www.npmjs.com/package/@motherduck/wasm-client) now supports `attach_mode='single'`, simplifying query execution and improving resource predictability when working with a single database. Refer to the [documentation](/sql-reference/wasm-client/) to learn more. - **Usernames added to Database listings in Settings:** MotherDuck Admins can now see the usernames for human users and service accounts on the Databases page in [Settings](/getting-started/interfaces/motherduck-quick-tour/#settings) for more intuitive lookups - **New export options for `EXPLAIN`:** MotherDuck notebook cells now support copying or exporting [`EXPLAIN` results](/sql-reference/motherduck-sql-reference/explain/) to simplify query inspection - **Enhanced Column Explorer experience for UUIDs:** The [Column Explorer](https://motherduck.com/blog/introducing-column-explorer/) now has added support for UUIDs and fields that default to top-N values for improved column-level insights and schema exploration. Refer to the [documentation](/getting-started/interfaces/motherduck-quick-tour/#column-explorer) to learn more. ## October 24, 2025 - **Duplicate MotherDuck notebook cells:** Duplicate cells in MotherDuck UI notebooks using the cell options menu or command menu. Access the duplicate option from the three-dot options menu on any cell, or use `Cmd/Ctrl + Shift + P` to open the command menu and search for "duplicate." ![Image](useBaseUrl('/img/about-motherduck/duplicate-notebook-cell.png')) ## October 9, 2025 MotherDuck now supports DuckDB versions 1.4.0 and 1.4.1, and DuckLake version 0.3 ๐ŸŽ‰ DuckDB 1.4 delivers performance gains with improvements like a rewritten sorting engine, more efficient small writes, and new SQL syntax including the MERGE statement. Learn more in the DuckDB [1.4.0](https://github.com/duckdb/duckdb/releases/tag/v1.4.0) and [1.4.1](https://github.com/duckdb/duckdb/releases/tag/v1.4.1) changelogs. ### Performance improvements - **[Sorting is 2x+ faster:](https://github.com/duckdb/duckdb/pull/17584)** Complete rewrite of sorting uses less memory and scales better across threads for ORDER BY, window functions, and list sorting - **[More efficient small writes:](https://github.com/duckdb/duckdb/pull/18829)** Appending small numbers of rows now writes far fewer bytes - **[5x faster checkpointing:](https://github.com/duckdb/duckdb/pull/18390)** Reuses table metadata when tables aren't altered during checkpoint - **[Parallel connection creation:](https://github.com/duckdb/duckdb/pull/18079)** Connections from instance cache can be created in parallel - **[Faster scalar functions on dictionary data:](https://github.com/duckdb/duckdb/pull/18127)** Functions on dictionary-compressed data only run once per unique value ### SQL syntax updates - **[`MERGE INTO` statement:](https://github.com/duckdb/duckdb/pull/18135)** Standard SQL upserts without requiring primary keys or indexes - **[`FILL()` window function:](https://duckdb.org/2025/09/16/announcing-duckdb-140.html#fill-window-function)** Interpolate missing values in ordered data - **[Python-style macro arguments:](https://github.com/duckdb/duckdb/pull/18684)** Macros accept positional or named arguments for any parameter - **[`STRUCT` to `MAP` cast:](https://github.com/duckdb/duckdb/pull/17799)** Direct casting between struct and map types ### Parquet improvements - **[`VARIANT` type reading:](https://github.com/duckdb/duckdb/pull/18187)** Read Parquet `VARIANT` types for faster semi-structured data processing - **[Native geometry type writes:](https://github.com/duckdb/duckdb/pull/18832)** Write native Parquet geometry types - **[Auto-globbing for directories:](https://github.com/duckdb/duckdb/pull/18760)** Automatically treats paths as directories and retries with glob patterns when no file is found Learn more in the official DuckDB Labs announcements for [1.4.0](https://duckdb.org/2025/09/16/announcing-duckdb-140.html) and [1.4.1](https://duckdb.org/2025/10/07/announcing-duckdb-141.html). While you can continue using your current version of DuckDB with MotherDuck, we encourage you to [upgrade your DuckDB clients to 1.4.1](https://duckdb.org/install) as soon as you can to take advantage of the fixes and performance improvements. ### [Preview] DuckLake 0.3 As we announced earlier this year, MotherDuck now supports [DuckLake](https://ducklake.select), an integrated data lake and catalog format. DuckLake 0.3 makes working with DuckLake more robust, including [`CHECKPOINT` for easy maintenance](https://github.com/duckdb/ducklake/pull/406), new paths for Iceberg interoperability, [spatial geometry types](https://github.com/duckdb/ducklake/pull/412), and [`MERGE INTO` support](https://github.com/duckdb/ducklake/pull/351). Learn more about using DuckLake databases in MotherDuck in the [documentation](/integrations/file-formats/ducklake), and the recent improvements in the [DuckDB Labs announcement for DuckLake 0.3](https://ducklake.select/2025/09/17/ducklake-03/). ## September 30, 2025 - **Get help from MotherDuck Experts:** Get a human helping hand with technical questions, troubleshooting, and best practices directly in the MotherDuck UI. Open "Expert help" from the Help menu to talk with our team, and you'll be notified of responses. Expert help is available with Business and Lite plans. - **Transient option for database storage retention:** Databases can now be created with transient retention, which provides a minimal retention period and no failsafe storage. This option can be useful for intermediate datasets or data easily reconstructed from external sources. Create transient databases in the UI or via [`CREATE DATABASE db_name (TRANSIENT)`](../../sql-reference/motherduck-sql-reference/create-database/#syntax). Transient databases are available with Business and Lite plans. Learn more in the [storage management documentation](/concepts/storage-lifecycle#storage-management). - **Duplicate notebooks:** Copy existing SQL notebooks to reuse query templates or create variations of your analysis. Find the duplicate option in any notebook's options menu in the left sidebar. - **Monitor database storage in the MotherDuck UI:** Organization admins can now review database storage metrics in the updated [Databases](https://app.motherduck.com/settings/databases) page, showing current and cumulative database storage footprint over time. Learn more in the [storage lifecycle documentation](/concepts/storage-lifecycle#breaking-down-storage-usage). ![Image](useBaseUrl('/img/about-motherduck/storage_monitor_ui.png')) ## September 10, 2025 - **Instances are now called Ducklings:** We've updated our name for instances to better reflect their purpose as dedicated and scalable DuckDB instances that provide isolated, on-demand compute for each user's analytics workload in MotherDuck. Find the familiar instance controls now in [Settings > Ducklings](https://app.motherduck.com/settings/ducklings). This release does not affect the [Admin REST API methods for instances](../../sql-reference/rest-api/motherduck-rest-api/). Learn more about how [Ducklings](../billing/duckling-sizes/) are different from standard data warehouse instances in [this blog post](https://motherduck.com/blog/scaling-duckdb-with-ducklings/). - **Rename Notebooks from the Object Explorer:** SQL notebooks can now be renamed directly from the left sidebar using a notebook's options menu. - **Enum support in `prompt` function:** The `PROMPT` SQL function now supports enum types for consistent classification outputs. See the [function documentation](../../sql-reference/motherduck-sql-reference/ai-functions/prompt/#classification-with-enums) for details and examples. - **Command menu in the MotherDuck UI:** Navigate the MotherDuck UI from your keyboard using the new command menu. Quickly access common actions like adding notebook cells, creating notebooks, and navigating between pages. Try it out with "Open command menu" in the top-left Organization dropdown, or use `Cmd/Ctrl + Shift + P` ## September 4, 2025 - **Pre-filled names for service accounts and tokens:** When creating service accounts and tokens in the [Settings > Service Accounts](/key-tasks/service-accounts-guide/manage-service-accounts-and-tokens/) page, names are now pre-filled with the following format to help differentiate between them: - _Service Accounts:_ `{creator_username}_service_account_{number}` - _Read-Write Tokens:_ `{sa_username}_read_write_token_{number}` - _Read-Scaling Tokens:_ `{sa_username}_read_scaling_token_{number}` - **DuckLake database icon in the MotherDuck UI:** [DuckLake-backed databases](/concepts/ducklake/) now display a distinct icon to easily distinguish them from databases using MotherDuck native storage. ## August 21, 2025 - **Support for H3 Spatial Indexing Extension:** MotherDuck now supports the [H3 DuckDB Extension](https://duckdb.org/community_extensions/extensions/h3.html), which adds support for the [H3 hierarchical hexagonal grid system](https://h3geo.org/) for geospatial analysis. This extension is pre-installed in MotherDuck, and users are not required to install this extension. ## August 13, 2025 - **GPT 5 Support in `prompt` function**: The `PROMPT` function now supports OpenAI's GPT 5 series models. Refer to the [function documentation](../../sql-reference/motherduck-sql-reference/ai-functions/prompt/) for more details. ## August 12, 2025 - **Display Preformatted VARCHAR values:** VARCHAR results in the MotherDuck UI data value pane now support display of preformatted text. - **Format SQL in MotherDuck Notebook:** Format any SQL statement using the new **Format** button in the notebook cell options menu, or with `Option/Alt + Cmd/Ctrl + O`. When text is selected, only the selection is formatted. ## August 8, 2025 - **Test S3 Credentials:** MotherDuck users can now test S3 credentials directly in the MotherDuck UI on the Secrets page in Settings when adding new S3 secrets. - **Support for DuckDB Configuration Options:** With this release, MotherDuck now correctly respects [DuckDB configuration options](https://duckdb.org/docs/stable/configuration/overview.html) and their local defaults, including extension settings like TimeZone. Broader coverage of additional configuration options is planned for the upcoming [DuckDB 1.4 release](https://duckdb.org/release_calendar.html). ## July 31, 2025 - **Updated FixIt Keyboard Shortcut:** The `Escape` key can now be used to reject [FixIt](/key-tasks/ai-and-motherduck/ai-features-in-ui/#automatically-fix-sql-errors-in-the-motherduck-ui) suggestions, providing a quicker way to dismiss generated SQL fixes. - **Generate Notebook Names:** Get descriptive, context-aware names for notebooks in the MotherDuck UI based on their SQL content. Click the new "Generate name from SQL" button to the left of a notebook's name to try it out. Available for users in MotherDuck's Business and Lite plans. ## July 25, 2025 - **Data Grid UX Improvements:** Data grids now include row numbers to make it easier to explore query results and reference specific rows. Users can now select multiple rows by clicking row numbers with the shift-key modifier. - **New UX for FixIt:** [FixIt](/key-tasks/ai-and-motherduck/ai-features-in-ui/#automatically-fix-sql-errors-in-the-motherduck-ui) now includes keybindings for the toggles to accept and reject suggestions and turn automatic suggestions on and off. - **`Cmd/Ctrl + Enter`:** Accept suggestion and run query - **`Cmd/Ctrl + Shift + Backspace`:** Reject suggestion ## July 16, 2025 - **NEW - Larger Compute Instances:** MotherDuck now offers two new memory-rich compute duckling (instance) types, **Mega** and **Giga**, built to run at high-capacity for the largest, most demanding jobs. Learn more in the [launch blog](https://motherduck.com/blog/announcing-mega-giga-instance-sizes-huge-scale) and [Docs](/about-motherduck/billing/duckling-sizes/). ## July 14, 2025 - **DuckDB 1.3.2:** MotherDuck supports DuckDB 1.3.2, a bugfix release. Additional details are available in the [DuckDB 1.3.2 changelog](https://github.com/duckdb/duckdb/releases/tag/v1.3.2). - **The Settings Button has Moved to the Org Dropdown:** Settings has moved from the left sidebar into the Organization dropdown at the top left for easier access and a cleaner layout. - **Admin Experience Enhancements:** With this week's release, MotherDuck organization admins now have the flexibility to do more to manage their Org directly from the MotherDuck UI due to better visibility and admin-specific functionality for managing tokens, service accounts, and storage. - **New Service Accounts Page in Settings:** Organization admins can now view, create, and manage service accounts and service account tokens in the [Service Accounts](/key-tasks/service-accounts-guide/manage-service-accounts-and-tokens/) section of MotherDuck settings. - **Impersonation of Service Accounts:** Organization admins can now temporarily [impersonate a service account](/key-tasks/service-accounts-guide/impersonate-service-accounts/) while using the MotherDuck UI. - **Storage Usage History added to `MD_INFORMATION_SCHEMA`:** Organization admins can now access up to 30 days of historical storage data using the [`STORAGE_INFO_HISTORY` view](/sql-reference/motherduck-sql-reference/md_information_schema/storage_info/) in the [`MD_INFORMATION_SCHEMA`](/docs/sql-reference/motherduck-sql-reference/md_information_schema/introduction/). Each record includes a `result_ts` timestamp showing when the storage metrics were calculated. ## July 01, 2025 **[Preview] DuckLake Support**: MotherDuck now supports [DuckLake](https://ducklake.select), an integrated data lake and catalog format. - MotherDuck currently provides two options for creating and integrating with DuckLake databases: - **Fully managed**: MotherDuck manages both data storage and metadata - **Bring your own bucket (BYOB)**: Connect your S3-compatible object storage with options for: - MotherDuck compute + MotherDuck catalog - Bring-your-own-compute (BYOC) + MotherDuck catalog Learn more in the [documentation](/integrations/file-formats/ducklake) and [announcement blog](https://motherduck.com/blog/announcing-ducklake-support-motherduck-preview/). ## June 26, 2025 - **Chat Widget Optimization:** Users can now view their inline edit history in a more compact chat widget and quickly request follow-up changes when needed. - ![Image](useBaseUrl('/img/about-motherduck/chat-widget-optimization.png')) - **Improved Boolean cell styling:** Boolean values in the data grid now have distinct visual weights to make it easier to visually scan result sets and prevent confusion with empty cells. ![Image](useBaseUrl('/img/about-motherduck/boolean-cell-styling.png')) ## June 18, 2025 - **DuckDB 1.3.1:** MotherDuck supports DuckDB 1.3.1, a bugfix release. Additional details are available in the [DuckDB 1.3.1 changelog](https://github.com/duckdb/duckdb/releases/tag/v1.3.1). - **`PIVOT` statements in MotherDuck UI:** The MotherDuck UI now supports [`PIVOT` statements](https://duckdb.org/docs/stable/sql/statements/pivot.html), with pivot columns also appearing in the Column Explorer. `PIVOT` transforms distinct column values into separate columns with aggregated data. - **New `STORAGE_INFO` View in `MD_INFORMATION_SCHEMA`:** Organization admins can now review detailed storage breakdowns per database using the new [`STORAGE_INFO` view](/sql-reference/motherduck-sql-reference/md_information_schema/storage_info/) in the [`MD_INFORMATION_SCHEMA`](/sql-reference/motherduck-sql-reference/md_information_schema/introduction/). ## June 12, 2025 - **Improved query execution UX:** After 5 seconds, the run button now displays a timer showing how long the query has been running. It also offers clearer visual cues for canceling a query on mouseover and focus. ## June 5, 2025 - **Overwrite a database with a zero-copy clone:** The new [`COPY FROM DATABASE (OVERWRITE)` command](/sql-reference/motherduck-sql-reference/copy-database-overwrite/) replaces all data in the target database with the source's contents in a single atomic operation, waiting for active writes to finish and blocking new ones during the process. - **Copy SQL definitions for views from the Object Explorer:** The dropdown menu for views in the left-hand panel of the MotherDuck UI now lets you copy the associated SQL definition without opening the table summary. ## May 29, 2025 MotherDuck now supports DuckDB version 1.3.0 ๐ŸŽ‰ DuckDB 1.3.0 improves performance in real-world scenarios for faster queries, new SQL syntax, and smarter Parquet file handling. Learn more in the [changelog](https://github.com/duckdb/duckdb/releases/tag/v1.3.0) here. ### Parquet improvements - **[New `TRY()` expression for safer queries:](https://duckdb.org/2025/05/21/announcing-duckdb-130.html#try-expression)** More graceful handling for bad data by returning `NULL` instead of an error on problematic rows - **[Pushdown of arbitrary expressions into scans:](https://github.com/duckdb/duckdb/pull/16430)** Reductions in unnecessary data processing to deliver up to 30x faster queries - **[Pushdown of inequality conditions into joins:](https://github.com/duckdb/duckdb/pull/16508)** Major speedups for incremental dbt models and join-heavy queries ### SQL syntax updates - **[Python-style lambda syntax:](https://github.com/duckdb/duckdb/pull/17235)** You can now use `lambda x: x + 1` instead of `x -> x + 1`; the old syntax is deprecated, but still supported. - **[`cast_to_type()` function:](https://github.com/duckdb/duckdb/pull/17209)** Dynamically cast values to match column types - useful in generic expressions and `CASE` statements when writing macros. - **[Recursive JSON access:](https://github.com/duckdb/duckdb/pull/17406)** New `json_each()` and `json_tree()` functions make it easier to traverse nested JSON structures. - **[Struct field updates:](https://github.com/duckdb/duckdb/pull/17003)** Individual fields in structs can now be modified using `ALTER`; all fields are rewritten even if only one is updated. - **[Prepared statements metadata:](https://github.com/duckdb/duckdb/pull/16541)** The `duckdb_prepared_statements()` function returns all prepared statements in the session. - **[More flexible type definitions:](https://github.com/duckdb/duckdb/pull/17404)** Support has been added for `CREATE OR REPLACE TYPE`, `CREATE TYPE IF NOT EXISTS`, and `CREATE TEMPORARY TYPE`. - **[Preserved order for `OR` filters:](https://github.com/duckdb/duckdb/pull/17180)** Execution now preserves the order of clauses in `WHERE` conditions using `OR`. - **[Function alias visibility:](https://github.com/duckdb/duckdb/pull/16600)** `duckdb_functions()` now returns aliases in addition to the function outputs. ### Parquet improvements - **[Late materialization:](https://github.com/duckdb/duckdb/pull/17036)** Queries are 3โ€“10x faster with `LIMIT` due to deferred column loading - **[~15% average speedup on reads:](https://github.com/duckdb/duckdb/pull/16595)** New scan and filter efficiency improvements - **[30%+ faster write throughput:](https://github.com/duckdb/duckdb/pull/17061)** Improved multithreaded export performance - **[Better compression for large strings:](https://github.com/duckdb/duckdb/pull/17164)** Large string values are now dictionary-compressed - **[Smarter rowgroup combining:](https://github.com/duckdb/duckdb/pull/17118)** Files are more efficient due to merging small rowgroups at write time Learn more in the official [DuckDB Labs 1.3.0 announcement](https://duckdb.org/2025/05/21/announcing-duckdb-130.html). While you can continue using your current version of DuckDB, we encourage you to [upgrade your DuckDB clients to 1.3.0](https://duckdb.org/docs/installation/?version=stable&environment=cli&platform=macos&download_method=package_manager) as soon as you can to take advantage of the fixes and performance improvements. ### Additional updates from this release are outlined below - - Query results now display in a redesigned table that delivers enhanced performance when viewing and exploring data - column headers now include type information for better context. Additional table functionality, including sorting and filtering of results, is coming in future releases. ## May 22, 2025 - **Faster queries on complex filters and wide tables:** We've significantly boosted performance for queries with IN filters, selective joins, and LIMIT clauses. Expect noticeable speedups on wide tables or those with large string or JSON columns. - **New keybindings for power users:** - Toggle Instant SQL for the current SQL cell: `cmd/ctrl+shift+.` - Toggle Object Explorer: `cmd/ctrl+b` - Toggle Inspector (Column Explorer): `cmd/ctrl+i` - Toggle worksheet mode for the current SQL cell: `cmd/ctrl+e` - **Org-wide Active Accounts:** Organization admins can now view all active accounts and their associated ducklings in the [Active Accounts](https://app.motherduck.com/settings/active-accounts) section of MotherDuck settings. - **Smarter Instant SQL caching:** Instant SQL now accounts for filters in your WHERE clause when building its cache, offering a greater number of relevant rows as you work. - **Full row count in flat table results:** SQL cells now display a full result row count when viewing results in "flat" table mode. - **GPT 4.1 Support in `prompt` function**: The `PROMPT` function now supports OpenAI's GPT 4.1 series models. Refer to the [function documentation](../../sql-reference/motherduck-sql-reference/ai-functions/prompt/) for more details. ## May 16, 2025 - **Multiple SQL statements now supported in Instant SQL:** Execute individual statements within multi-statement SQL cells by clicking on the desired statement while [Instant SQL](https://motherduck.com/blog/introducing-instant-sql/) is enabled. - **Copy Table Names directly from Object Explorer:** Use the options menu on any table in the Object Explorer to copy its name to your clipboard. Paste exact table references into any SQL editorโ€”eliminating typos and saving time when writing queries. For earlier updates, see the [release notes archive](/about-motherduck/release-notes-archive/). --- ## Docs feedback MotherDuck accepts optional user-submitted feedback about this page at `POST https://motherduck.com/docs/api/feedback/agent`. For agents and automated tools, feedback submission should be user-confirmed before sending. Payload: ```json { "page_path": "/about-motherduck/release-notes/", "page_title": "Release notes", "text": "", "source": "" } ``` `page_path` and `text` are required; `page_title` and `source` are optional. Responses: `200 {"feedback_id": ""}`, `400` for malformed payloads, and `429` when rate-limited.