SlashDB ver. 2.2 (Summer 2026)

SlashDB version 2.2 is now available for download and for instant launch on Azure, AWS, and Docker-based environments.

This release delivers a range of practical improvements across the product, including DuckDB support for analytical workloads, an enhanced Data Discovery experience, streamlined local installation and startup workflows for developers, and continued gains in performance, diagnostics, and usability. In total, this release includes over 95 work items.

SlashDB architecture also continues to evolve. Ongoing refactoring efforts are making more features optionally installable as plug-ins, helping keep the core platform lean, modular, and easier to extend.

For upgrade instructions please refer to Upgrading chapter in the User Guide.

What’s new in version 2.2?

🔍 Advanced filtering in Data Discovery

Data Discovery now offers a significantly more powerful filtering experience through a new filtering panel and expanded modifier controls. Users can access more of SlashDB’s API filtering capabilities directly from the web interface, making it easier to refine datasets without manually building request paths. 

The new experience is designed to make advanced filtering more discoverable and more practical in day-to-day work. It supports:

• Column match to value (whole and partial)
• Logical (AND, OR, NOT) for precise multi-value filtering
• Traversal across related tables while preserving filtering logic
• Column selection within result sets
• Result modifiers, including distinct, streaming, transposing and record count

Table headers now feature a context menu for hiding columns, adjusting sort order and opening advanced filtering.

🦆 DuckDB support for analytical workdloads

SlashDB 2.2 adds DuckDB as a fully supported database backend. This opens the door to new analytical and embedded use cases built on one of today’s most popular open-source analytical databases. Teams can now expose DuckDB-based data through SlashDB’s API and web interface just as they do with more traditional relational backends.

📦 Local installation on Windows

Developers and data engineers can now install SlashDB with standard pip workflows on Windows workstations, include only the database and feature dependencies they need, and use the new init and serve commands to get local instances up and running more quickly.

These packaging and CLI improvements make SlashDB easier to evaluate, faster to prototype with, and simpler to integrate into modern Python-based development environments. Below is an example of launching SlashDB on Windows with support for Microsoft SQL Server and PostgreSQL databases in just a few keystrokes.

PS D:\> python -m venv sdbwin
PS D:\> .\sdbwin\Scripts\activate
(sdbwin) PS D:\> pip install --find-links https://downloads.slashdb.com/latest slashdb[mssql,postgresql]
Looking in links: https://downloads.slashdb.com/latest
Collecting slashdb[mssql,postgresql]

[...]

Successfully installed alchemyjsonschema-0.8.0 attrs-25.4.0 beaker-1.13.0 boltons-25.0.0 certifi-2025.11.12 cffi-2.0.0 charset-normalizer-2.1.1 click-8.3.3 colorama-0.4.6 concurrent-log-handler-0.9.29 cryptography-48.0.0 dictknife-0.14.1 greenlet-3.2.4 hupper-1.12.1 idna-3.11 isodate-0.7.2 jsonschema-4.26.0 jsonschema-specifications-2025.9.1 jwcrypto-1.5.7 lxml-6.0.2 lz4-4.4.5 magicalimport-0.9.1 mako-1.3.10 markupsafe-3.0.3 packaging-26.2 passlib-1.7.4 paste-3.10.1 pastedeploy-3.1.0 pastescript-3.7.0 plaster-1.1.2 plaster-pastedeploy-1.0.1 platformdirs-4.9.6 portalocker-3.2.0 psycopg2-2.9.12 pycparser-2.23 pyjwt-2.12.1 pyodbc-5.3.0 pyramid-1.10.8 pyramid-beaker-0.9 pyramid-exclog-1.1 pyramid-mako-1.1.0 pyramid-multiauth-0.9.0 python-dateutil-2.9.0.post0 python-memcached-1.62 pytz-2025.2 pywin32-311 pyyaml-6.0.3 referencing-0.36.2 requests-2.34.2 rpds-py-0.27.1 setuptools-80.9.0 simplejson-4.1.1 six-1.17.0 slashdb-2.2.0 slashdb-mssql-1.2.0 slashdb-postgresql-1.2.0 sqlalchemy-2.0.49 sqlparse-0.5.5 strict-rfc3339-0.7 translationstring-1.4 typing-extensions-4.15.0 urllib3-1.26.20 venusian-3.1.1 waitress-3.0.2 webob-1.8.9 wsgicors-0.7.0 zope-deprecation-6.0 zope-interface-8.0.1 zxcvbn-4.4.28

(sdbwin) PS D:\> slashdb init
Directory in which to set up SlashDB [C:\Users\volx\AppData\Local\SlashDB]: d:\sdbwin\etc
Copying files to d:\sdbwin\etc
Replacing templates.
SlashDB initialized at d:\sdbwin\etc
Run `slashdb-serve --ini d:\sdbwin\etc\slashdb.ini` to start SlashDB

(sdbwin) PS D:\> slashdb-serve --ini d:\sdbwin\etc\slashdb.ini
Starting SlashDB from INI at d:\sdbwin\etc\slashdb.ini
Serving at http://0.0.0.0:6543
Starting server in PID 51908.

🤖 AI assistant extension proof of concept

SlashDB’s extension framework was also used to demonstrate a proof‑of‑concept AI integration, highlighting the types of client‑specific customizations the platform can support. The demos were very well received and included integrations with Google Gemini, Ollama, Microsoft Copilot, and OpenAI ChatGPT.

The extension framework is part of core SlashDB, while the AI integration itself is an example of what can be built on top of it. If this is of interest for your environment, please contact us if you would like to see a demo.

AI agents were introduced with new capabilities to generate new query definitions from natural language prompts, as well as to explain existing queries in natural language. Users remain in full control, with the ability to review, accept, or modify generated queries, which can then be instantly exposed as API endpoints.

🔢 Record count toggle and improved pagination in user interface

Data Discovery now exposes the record count feature in the web interface and automatically calculates the number of pages.

⚠️ Behavior changes and deprecations

Static content

We are changing how certain static content is served. If you are upgrading from prior versions please make sure to edit the following.

From nginx.conf remove this section entirely.

In slashdb.ini you need to remove the static_path variable or edit the as follows:

static_path = slashdb.views:static

You can also remove the original location for the static_path from the file system (i.e. /var/lib/slashdb/www), if it still exists after the upgrade.

Python 3.9

Python 3.9 is no longer supported. Supported versions are now 3.10, 3.11, and 3.12.

🛠️ Additional improvements and fixes

• Schema for transposed output rounds out schema enhancements from last release
• Pagination in the UI is improved with record count visibility and direct page navigation.
• Server-side cursors and streaming controls are more flexible and more explicit in the web UI.
• Data Discovery and SQL Pass-thru usability have been improved for very large schemas, including better performance and table search ergonomics.
• Database connection speed is faster for cached schemas and improved for uncached schemas.
• The slashdb-diag toolkit is more efficient on very large log files, with improved handling of diagnostic output.
• Request Status has improved detection of interrupted or externally failed requests, and refelect status polling is less noisy.
• Databricks timeout, invalid token and shutdown scenarios now surface more meaningful behavior and error handling.
• SQL Pass-thru and Data Discovery handle more edge cases correctly, including transposed output behavior, special characters in filters and parameters, vectors and arrays in HTML output, and column selection URL handling.
• Validation and error messages are clearer in several workflows, including date and time fields, missing mapped database credentials, invalid query loading, and license replacement when the current license is invalid.
• A range of Windows-specific fixes improve configuration saving, SQLite connectivity, and local installation reliability.
• Numerous UI and API bug fixes improve consistency across configuration, filtering, data entry, and permissions workflows.
• Security-related fixes, including defensive measures against Linux kernel vulnerabilities CopyFail, DirtyFrag/Fragnesia and dependency updates.

💬 Connect with us

We invite you to get in touch with us. You can reach us via:

• StackOverflow (use tag “slashdb” to post questions)
• Contact form

🖼️ About the header image

The image ws generated by Google AI with Nano Banana 2. Actual SlashDB content was used in making of the image.