Agentic applications and orchestration

Mendix Cloud GenAI: Multi-model access and flexible resource capacity

Multi-model access, simplified

Text generation resources now give you access to multiple Claude model families (Haiku, Sonnet, and Opus) from a single resource with one API key. Choose the model families you want when provisioning, and they’re immediately available. As Mendix introduces new model versions, they’re automatically added to your existing resource. No manual updates, no new keys—just seamless access to the latest capabilities.

Flexible capacity that scales with you

Fixed plan sizes are gone. Text generation and embedding resources now use a flexible capacity model based on GenAI Units, where one Cloud Token = 100 GenAI Units per month. Each model family consumes units at different rates, so you can match the right model to each task; lighter models like Haiku for simpler workloads, more powerful models like Sonnet and Opus when you need them, all from the same monthly allocation. Scale up instantly, optimize as you go.

Getting started

Upgrade to Mendix Cloud Connector v6.2.0 or higher to access multi-model selection in your applications. Set a default model per resource to maintain compatibility with apps on older connector versions.

Agents Kit 2.0

With Mendix Studio Pro LTS 11.12, we’re also launching Agents Kit 2.0, a major milestone in how you build AI agents with Mendix.
Starting with 2.0, the Agent Editor is the preferred way to build agents. It lets you define agents at design time directly in Studio Pro, manage their full lifecycle as part of your app model, and leverage built-in platform capabilities like Model documents, version control, and deployment pipelines. Develop locally, deploy to the cloud without any friction.

And to keep pace with the fast-moving AI landscape, Agents Kit will now ship monthly updates, so you always have access to the latest capabilities to build maintainable, scalable agentic applications including updates and additions to our existing components for:

Connectivity

• Exposing tools with MCP (Model Context Protocol) servers built in Mendix
• Consuming (external) MCP tools with a Mendix MCP Client
• Minimal configuration effort when accessing generative AI via Mendix Cloud GenAI
• Support of most common GenAI providers such as Amazon Bedrock, (Azure) OpenAI, Mistral, and Google Gemini
• Bring-your-own-LLM: Connect your own language model for maximum flexibility and control

Observability

• AI Observability via Traces and Human-in-the-Loop controls including tool call visibility and human approval steps in order for you to gain full transparency and control over AI behavior

Developer experience

• Reusable Conversational UI snippets

Getting started with Agents Kit: New onboarding experience and module compatibility overview

The Agent Editor onboarding experience now guides developers step-by-step with essential documentation to quickly set up their first AI agent in Mendix and integrate it seamlessly into existing application logic.

Keeping track of the dependencies of our Agents Kit and their frequent updates for new features can be a real headache – but not anymore! We’re thrilled to introduce the Agent Editor’s new compatibility table, offering a clear overview of module dependencies and version compatibility, complete with direct links to download compatible versions from our marketplace.

Native Support of the Agent Editor in showcase and starter apps

Within days after the 11.12 release, you will be able to start a new app with the Blank GenAI or Agent Builder Starter App, or to explore what’s possible with the Agent Editor by downloading the GenAI showcase application or the RFP Assistant Starter App in 11.12. The latter is then fully powered by Studio Pro agents, replacing the previous approach of orchestrating logic through microflows with manual LLM calls, prompts, RAG, and function calling.

Note: Mendix Cloud GenAI is the supported provider for this app at launch.

Upgrade existing applications with the latest versions of our GenAI components in 11.12 to make use of all the latest features. And keep your eyes open for new Academy content for agentic applications!

Agentic development

Maia Plan: Push your plan straight into Jira

Maia Plan now connects to Jira, so the epics and user stories Maia generates land directly in the backlog your team already works from, no need to switch to the Mendix Planning Board to start delivery. If Jira is where your team plans, refines, and runs sprints, Maia Plan now meets you there.

In the Start Development step, choose Jira as your planning tool and connect it once with your Jira environment URL, project key, board ID, account, and API token. Maia pushes the full scope into your Jira backlog, grouped by epic, ready to prioritize. Pull the stories you want to tackle into an active sprint, and that scope becomes available in Studio Pro, where Maia Make picks it up and starts implementing. Prefer the Mendix Planning Board? It’s still the default, switchable anytime in project settings.

The result: from idea to a prioritized, execution-ready backlog in the tool your team already lives in.

Maia Make: module skills

Skills in Maia Make are no longer limited to your own app. Any module — including modules you install from the Marketplace — can now ship its own set of skills. This makes it straightforward to share specialized prompts and patterns across applications: publish a module, and its skills travel with it. Teams building reusable building blocks for their organization can now include Maia skills as a first-class part of the package.

The skills overview dialog shows you a list of all your skills in your project with their respective status, so you can validate that they are loaded correctly.

Paste images into the chat

You can now paste images directly into the Maia Make chat window, in addition to the existing file upload flow. Grab a screenshot, a wireframe, or a snippet from your browser and paste it in-place; no need to save the file first. It’s a small change that removes a surprising amount of friction from the day-to-day workflow.

Plan mode, clarifying questions, and improved reasoning

Maia Make now has an improved plan mode that runs before it starts generating. If your prompt is ambiguous or underspecified, Maia will ask clarifying questions rather than making assumptions and producing something you’ll immediately want to undo. The underlying reasoning has also been improved in this release, with Maia better able to break down complex requirements, consider trade-offs, and produce more coherent multi-step plans before writing a single line of model. The result is fewer wasted iterations when the task is non-trivial.

Add Markdown documents to your prompt

Alongside PDFs and images, Maia Chat now accepts Markdown files as context. Upload an architecture document, a specification, an API reference, or any other Markdown content, and Maia will work from it directly. This rounds out the document types you can use to seed a conversation — useful when your team’s documentation is already in Markdown (as it increasingly tends to be).

Thank you Patrick Bornier for this suggestion.

MCP server and client improvements

The Studio Pro MCP server is more stable in 11.12, and the built-in MCP client now supports a richer set of authentication options, including improved bearer token authentication. If you connect external agents or tools to Studio Pro via MCP, you should notice the agent has better instructions how to use the MCP server, and improved stability when calling the MCP tools in Studio Pro. No configuration changes are needed to pick up the stability improvements.

Bring your own LLM

You can now configure a custom language model backend for Maia Make in Studio Pro. Any OpenAI-compatible API is supported, which covers locally-hosted models, models accessed via OpenRouter, and self-managed inference endpoints, in addition to OpenAI itself. This gives you control over cost, data residency, and model choice without losing Maia’s IDE integration, skills, and Mendix knowledge base access. Configuration is done in Studio Pro’s preferences; no other tooling is required.

Purchase additional Maia Units using Cloud Tokens

For those who use Maia Make intensively — beyond what’s included with your Mendix plan — Mendix 11.12 introduces the option to purchase additional Maia Units using Cloud Tokens. If you find yourself running low, simply convert Cloud Tokens you already have available rather than hitting a hard stop mid-project. Your Mendix Administrator can help if you need additional Cloud Tokens to top up with.

This is an initial step toward giving customers and developers more direct control over their Maia consumption. More options for managing and extending usage will follow in future releases.

Marketplace module search and download

Maia Make can now search the Mendix Marketplace and download modules directly into your project. Rather than manually searching and downloading it, you can describe what you need in the chat and let Maia handle the lookup and installation. This is particularly useful mid-task: if Maia determines that a module would be the right building block for what you’re asking for, it can fetch it without interrupting your flow.

Broader document type support in Maia and MCP

Maia Make and the Studio Pro MCP server can now read and edit a wider range of document types. New in this release: non-persistent entities in the domain model, data transformations (JSLT), and three additional workflow elements — AI Tasks, Event Subprocesses, and Message Events. If your app relies on any of these, Maia can now work with them directly rather than stopping short and asking you to handle them manually. The same applies to external agents connecting via MCP: the expanded surface area is available to any MCP-capable tool.

Unit Tests and Validation microflows

Maia Make is now more closely aligned with best practices for both Unit Test microflows and Validation microflows as part of its output. Ask Maia to build a piece of logic, and you can now ask it to test and validate that logic in the same conversation. This brings automated quality checks into the generation workflow rather than treating them as a separate, manual follow-up step — useful both for teams with existing test coverage standards and for anyone who wants a quick sanity check before committing generated logic to the model.

Full widget and design property support

For pages, Maia Make now works with all pluggable widgets and supports setting design properties, including custom ones from your own design system. Previously limited to a hardcoded set of widgets, Maia can now generate pages that make full use of your widget library and reflect your design system, supporting a much broader range of real-world applications.

Widgets

Dynamic tree hierarchies with Tree Node

The Tree Node widget (part of the Data Widgets module) now supports dynamic recursive hierarchies. Previously, tree structures required you to define each level of the hierarchy statically in Studio Pro, making it impractical to work with data of unknown or variable depth. By configuring a parent association on your datasource, the widget now automatically builds the tree from your data, no matter how deep it goes. This makes it easy to build things like org charts, category trees, or file structures.

Signature Widget is now React compatible

The Signature widget has been rebuilt for the React client, ensuring compatibility with future Mendix versions. It retains all its existing capabilities: capturing signatures on a canvas, customizable pen color and size, a toggleable clear function, and storing the result as an image via the Image widget. Note that upgrading requires a one-time migration.

PWA Wrapper now generally available

Progressive Web Apps are one of the fastest ways to build mobile applications, and with the PWA Wrapper we are extending that approach to cover the vast majority of mobile app use cases for Mendix customers — targeting up to 90% of typical mobile requirements. It allows you to reuse your existing web pages, styling, and logic while benefiting from the flexibility and performance of modern web technologies, and now also the reach of native distribution.

Historically, PWAs have had two key limitations:

1. They cannot be distributed through the Apple App Store or Google Play Store as native applications, and
2. They are limited to the capabilities exposed through web APIs

The PWA Wrapper addresses both. It packages your Mendix PWA as a native iOS and Android application, making it eligible for distribution through the app stores in a way that aligns with platform review guidelines (including Apple’s requirement that apps provide value beyond being a simple website wrapper). At the same time, it provides a curated set of native device capabilities that covers the needs of most enterprise mobile applications, reducing the need for custom native modules or separate build pipelines.

The result is a new way to build mobile apps: faster, simpler, and without compromise. By combining the speed of PWAs with the capabilities of native applications, the PWA Wrapper eliminates the traditional trade-offs of mobile development. For most Mendix projects, it’s not just another option — it’s the new default

React Native upgrade

React Native is the technology that powers Mendix native mobile apps on iOS and Android. To maintain compatibility with the latest iOS and Android OS versions and stay current with the open-source ecosystem, Mendix regularly upgrades React Native. Mendix 11.12 makes a significant jump to React Native 0.84, which introduces a fundamentally redesigned architecture that removes the legacy bridge between JavaScript and native code. This architectural shift delivers faster startup times, quicker iOS builds, improved rendering performance, and a smaller app footprint.

Mendix has updated the Native Template and all platform-supported modules to work with the new React Native version. Several underlying libraries changed substantially in the process, including the push notification library, the splash screen library, the image caching library, and the animation library. Developers building custom widgets and JavaScript actions will also benefit from faster Hot Module Replacement during development, clearer error messages from the code generator, and improved debugging tools with React DevTools 5.

All customers currently on Mendix 10.24 LTS should upgrade to 11.12 to maintain continued support for their native mobile apps. Without this upgrade, we cannot guarantee that apps will continue to work correctly on upcoming iOS and Android OS releases. If your app uses custom widgets, JavaScript actions, or any other custom native code or modules, verify that they are compatible with React Native 0.84 and React 19. The Native Template shows which versions are in use today.

Automatic code migration during updates

Mendix regularly evolves its runtime and client APIs to improve performance, security, and extensibility. When updating, Java actions and JavaScript code that use older APIs may need to be updated to stay compatible. In larger applications, identifying and applying those changes across dozens of files can be time-consuming work.

From Mendix 11.12 on, Studio Pro can automatically migrate both Java and JavaScript code to address API deprecations and planned runtime and client API changes. The recommended workflow is to fix deprecations before updating: open the Update Assistant in your current version of Studio Pro (View > Update Assistant), scan your app, review the detected issues, and apply the suggested fixes.

For customers updating from Mendix 10.24 LTS, the Update Assistant is not available in Studio Pro 10. To make sure you can still prepare your code before updating, we are releasing standalone CLI tools that work with your 10.24 project. For guidance on how to use them, see our documentation.

Richer observability: OpenTelemetry logs and improved trace names

Traces tell you where time is being spent in your application; logs tell you what was happening at that moment. Used together, they are one of the most powerful tools for diagnosing production issues — but getting both flowing through OpenTelemetry in Mendix required separate configuration steps in different places, making it easy to end up with tracing set up but logs going elsewhere. And even when tracing was set up, runtime operation traces only showed the type of operation — retrieve, delete, and so on — without indicating which entity or XPath expression was involved. That forced teams to cross-reference trace data with application logic manually.

Mendix 11.12 delivers improvements on both fronts. Studio Pro now has a dedicated OpenTelemetry setting where you configure your endpoint once and then simply enable tracing, logs, or both — no more hunting through separate configuration screens. At the same time, runtime operation span names now include the full XPath expression, so it is immediately clear which data operation a trace refers to — no cross-referencing required.

To get started with log export, open the OpenTelemetry tab in your app’s runtime settings in Studio Pro and enable the logs option. For more information on tracing and log configuration, see Tracing in the Mendix documentation.

Studio Pro performance improvements

Studio Pro is where Mendix developers spend most of their working day, and performance directly impacts how quickly you can build, test, and iterate. As applications grow larger and models become more complex, performance becomes increasingly important. Small improvements across common operations — project loading, consistency checks, local deployments — compound over the course of a working day into meaningful productivity gains.

Mendix 11.12 brings substantial performance improvements that make Studio Pro noticeably faster throughout your working day.

• Project load times are 25-40% faster from your very first open
• Error checking runs up to 6x faster on large projects thanks to parallel consistency checks on multi-core machines

Local deployment sees significant gains during active development sessions:

• After your first deployment, subsequent deployments in the same session are up to 8x faster, substantially shortening the test-and-iterate cycle

Studio Pro also manages memory more efficiently, reducing performance degradation over time — particularly beneficial when working long sessions throughout the day.

• For developers on Mac, Studio Pro 11.12 runs an average of 35% faster thanks to native Apple Silicon builds, making this the fastest Mendix development environment we’ve ever shipped

Studio Pro 11.12 requires Apple Silicon and no longer supports Intel-based Macs.

These improvements are active immediately when you install Studio Pro 11.12. No configuration changes required. You’ll feel the difference from the moment you open your first project.

Find Results gets a boost: Faster navigation with built-in search history (beta)

Searching your model is something developers do constantly, and Studio Pro’s Find Results panel is the gateway to rapid navigation across large applications. Switching between searches or revisiting earlier queries breaks the flow and wastes time.

Mendix 11.12 transforms the Find Results panel with a redesigned interface and built-in search history. You can now instantly switch between past and current results without re-running searches or managing multiple panels. The result is faster, smarter navigation that keeps you focused on building.

Enable the new find panel in Studio Pro preferences to unlock this improved workflow.

Embed Mendix inside any web application (beta)

Integrating Mendix capabilities into an existing portal, product, or custom frontend has traditionally meant reaching for an iframe. Iframes work, but customers consistently find them limiting — the integration never feels seamless, and delivering a polished, cohesive experience requires workarounds that add complexity and maintenance overhead.

Mendix 11.12 introduces the Embedded Client (public beta), which lets you mount a Mendix web app as a web component inside any host application — React, Vue, plain JavaScript, or any other frontend framework — without an iframe. The host application owns the surrounding page and browser experience; Mendix owns the region where it is mounted. Styles are isolated via a shadow root, Atlas styling is fully supported, and the embedded app can receive page parameters from the host at mount time. Authentication in the beta requires anonymous users; support for passing authenticated sessions from the host is planned for a future release.

To get started, add an Embedded navigation profile to your app in Studio Pro (App > Navigation > Add navigation profile > Embedded), deploy, then load the embedded bundle from your host application with a dynamic import. For full integration instructions, code examples, and guidance on CORS and CSP configuration, see Embedding the Client in the Mendix documentation.

Broader support for Java action optional parameters

We’ve expanded the types of parameters that can now be optional in Java actions to include Entity and Microflow – streamlining your development process.

This enhancement helps you avoid common workarounds, such as:

• Selecting a dummy microflow when no specific microflow is needed.
• Adding extra actions to pass objects to microflows, which previously required additional development steps.

By allowing these Java action parameters to be optional, we aim to reduce complexity and save you development time.

Comparing multiple historical revisions

With 11.9 we introduced comparing a historical revision with the current state of your app to quickly see how your app has changed. With this release, we’re excited to announce any-to-any diffing. Select multiple commits from the History pane to start a comparison on the Comparison pane and inspect a detailed textual overview of all modifications, like the Changes pane for uncommitted work.

And what’s next? We’re switching gears to Visual Diffing to allow showing multiple versions of, for example, a microflow side-by-side.

Domain Model Editor improvements

In 11.11 we introduced the refreshed look and feel of the domain model editor, giving it a more modern and polished experience. In 11.12, we are building on that update with a small but useful addition: a legend.

The new legend helps clarify the different visual elements in the domain model editor, making it easier to understand at a glance and improving overall learnability, especially for newer developers. It is a small change that makes the editor more approachable and easier to work with.

Variables pane maintains state and display

The Variables pane has been updated for a smoother debugging experience. The variable pane now maintains its state and display during debugging operations (such as Step Over, Step Into, or Continue), ensuring that your variables and their values remain visible and accessible without collapsing or resetting unexpectedly.

Manage Mendix Cloud with AI using MCP

AI is rapidly becoming part of how developers build and operate applications. With the new MCP integration for Mendix Cloud, you can now connect AI coding assistants such as Claude Code and GitHub Copilot directly to your Mendix Cloud environments. This allows you to interact with your apps and environments using natural language, instead of manually calling APIs or navigating the Mendix Portal. Whether it’s listing your apps, deploying a package, or managing environments, day-to-day operations become faster and more intuitive.

The integration exposes Mendix Cloud APIs as tools that AI agents can understand and use natively. Once connected, your assistant can perform a wide range of tasks across the application lifecycle, including managing environments, scaling resources, deploying packages, and checking deployment status. This helps reduce the effort of repetitive tasks, streamlines operational workflows, and removes the friction of working directly with APIs.

At the same time, the MCP integration is designed to fit seamlessly into your existing setup. You can connect using tools you already use, such as Claude Code, GitHub Copilot, VS Code with Copilot, or other MCP-compatible clients. This ensures you stay in control while extending your development and operations workflows with AI-driven capabilities.

By connecting Mendix Cloud to AI assistants through MCP, you unlock a more efficient and flexible way to manage your applications. From simplifying deployments to enabling automation through natural language, this integration helps you focus more on building and improving your applications, and less on operational overhead.

Enterprise-grade platform

Track consumption accurately, optimize licensing costs with End User Telemetry

As a Mendix Platform Administrator, you want complete visibility into user licenses and usage of all your apps.

With End User Telemetry, platform administrators can seamlessly track active named users across all production apps from a single, centralized dashboard. No more time-consuming spreadsheets or guessing games – user telemetry runs continuously in the background to give you precise, aggregated data on your actual user license consumption.

To protect data privacy, all user identities are securely hashed at the source before aggregation. Usage is categorized as per the commercial model of:

• Single-App Internal Users
• Multi-App Internal Users
• External Users

You can easily track monthly trends or view daily usage snapshots. User Metering is smart and fail-safe by design: it automatically deduplicates cross-app users. It is also entirely risk-free; if your apps experience a sudden spike and exceed seat limits, there is absolutely zero app disruption – your apps keep running allowing you to manage your licenses smoothly.

With User Metering, license governance becomes proactive, transparent, and completely automated so you can scale your platform ecosystem with financial confidence. Check out our documentation for more details.

Define rules once, apply everywhere with Mendix Policies

As a Mendix Platform Administrator you want to make sure your guardrails are always applied.

With Policies, platform administrators can define governance rules once and automatically apply them across every app. No more time‑consuming reviews or reliance on external tools—policies run continuously during development and deployment to help keep applications secure and compliant.

Start quickly with ready‑to‑use templates or create your own. Today, you can define policies for:

• Mendix runtime version
• Marketplace Licenses Type
• Marketplace Support Status
• Software Composition Findings

You can easily track violations per app and drill down into detailed results.

Policies are flexible by design: target specific apps or your entire portfolio and expand as new policy types are introduced. Policies are also deterministic; no AI is involved.

This is just the beginning; policies now bring visibility, but no enforcement (that will follow soon along with additional policy types). With Policies, governance becomes proactive, scalable, and built into the development flow so you can grow your platform with confidence. See our introduction video for more details.

Visibility for known vulnerabilities in Software Composition

We’ve enhanced Software Composition to give you instant visibility into known Mendix/Siemens vulnerabilities as reported in the Security Advisories across your Mendix applications. Whenever a Mendix or Siemens-supported component contains a known issue, it’s automatically surfaced—clearly prioritized from critical to low based on CVSS scores.

Easily filter per app to see exactly where risks exist, and understand what to do next with clear, actionable Mendix guidance. Stay ahead of vulnerabilities, focus on what matters most, and keep your applications secure—without slowing down delivery.

Note that because of this new feature, the supported Mendix versions for Software Composition have changed. If your application is an unsupported version, this is indicated in software composition with a warning icon.

Publish Marketplace components straight from your CI/CD pipeline

The new Marketplace Publish API lets you manage the entire release lifecycle of your components through code. You can create a new release with its version metadata, upload the source files, track the release as it moves through review and publication, and unpublish it later if needed — all without opening the Marketplace portal. It works for both public and private components.

The real benefit is automation. If you build components and keep your source in tools like GitHub, GitLab, or Azure DevOps, you can now make publishing part of the same pipeline you already use to build and test. That removes manual, error-prone steps, makes every release repeatable, and lets you ship updates as often as your pipeline runs. Your release process becomes consistent and auditable instead of a series of hand-done actions.

Create multiple component drafts and review them together

You can now create and manage multiple component drafts at the same time. You no longer have to finish and publish one draft before starting the next, so teams supporting Studio Pro 9, 10, and 11 side by side can prepare and ship component updates for each version on its own schedule, without the old one-draft-at-a-time bottleneck.

Drafts are also no longer a solo activity. Colleagues listed as developers on a component can see each other’s drafts, so a teammate can review your files, description, and screenshots before anything goes live — no exporting or sharing files outside the platform. Visibility stays controlled: only people assigned as developers on that component or are a member in the component group, can see its drafts, so work in progress stays within the team.

Change Data Capture (beta): stream app data to data lakes in real time

Modern businesses demand instant access to accurate data for informed decision-making. In Mendix 11.12, we’re introducing Change Data Capture (CDC) in beta: a powerful capability that streams data changes from your Mendix application to data lakes in a simple and more performant way. Instead of pulling large amounts of data regularly via REST or OData queries that can impact the performance of your Mendix app, CDC pushes changes as they happen.

You can now capture every create, update, and delete operation on your domain model entities and stream them in real-time to destinations such as Amazon S3 or Azure Data Lake. Simply select which entities to track in Studio Pro, configure your Event Broker connection, and set up the Iceberg bridge. Your data change events will be transformed into Iceberg tables and pushed into the data lake. This opens powerful use-cases like enabling real-time business intelligence dashboards, maintaining comprehensive audit trails, or feeding data into AI pipelines.

Change Data Capture is available as a beta feature in Mendix 11.12. Event broker and Iceberg bridge are available for deployments on Mendix Cloud.

Data Transformer is now generally available

Following its beta release in 11.11, Data Transformer is now generally available in Studio Pro 11.12. Data Transformer helps developers map and transform structured data more easily, making it simpler to work with external integrations and data processing scenarios directly within Studio Pro. Since its beta release, the feature has remained stable in both functionality and user experience.

What’s new in 11.12 is Maia support for Data Transformer. You can now ask Maia to help create Data Transformer and generate JSLT expressions, making it faster and easier to create transformations. This Maia capability significantly lowers the barrier to working with JSLT and can speed up the process of building integrations.

With general availability in 11.12, Data Transformer is ready for production use. Learn more about setting up Data Transformer or how Data Transformer helps simplify your integration.

Improve application observability with OpenTelemetry

Gaining insight into what happens inside your application in production is essential for maintaining performance and reliability. With the introduction of OpenTelemetry support in Mendix, you can now trace requests end-to-end across your application, from microflows to external dependencies. This gives you a clear understanding of where time is spent, how requests flow through your app, and where issues originate, helping you diagnose and resolve problems faster.

Because this capability is built on OpenTelemetry, it integrates easily with the observability tools you already use. You can correlate traces with logs and metrics to get a complete picture of your application’s behavior, without being tied to a specific vendor. This allows you to strengthen your monitoring setup, improve performance tuning, and operate your Mendix applications with greater confidence.

STACKIT Kubernetes Engine support in Mendix on Kubernetes

Multi-cloud flexibility meets European sovereignty

Mendix on Kubernetes now supports STACKIT Kubernetes Engine (SKE) as its fourth validated Kubernetes flavor, joining AWS EKS, Azure AKS, and Google GKE. STACKIT is the European sovereign cloud operated by Schwarz Digits (the IT division of Germany’s Schwarz Group), with data centers exclusively in the EU.

For European organizations — particularly in the public sector, financial services, and other regulated industries — this removes a critical sovereignty barrier. SKE gives you the same Mendix on Kubernetes deployment experience you’re already familiar with, but on infrastructure that’s provably under EU jurisdiction and immune to extra-territorial US legislation like the CLOUD Act.

What you get

• True data sovereignty: EU-headquartered provider under EU jurisdiction, with no exposure to non-EU legal frameworks
• Compliance built in: STACKIT carries BSI C5, ISO 27001, GDPR certification, and alignment with emerging standards like SecNumCloud and DORA
• No new learning curve: SKE uses the same Mendix on Kubernetes deployment model; it’s just another validated flavor
• Industrial-strength trust: backed by Schwarz Group (parent of Lidl and Kaufland), with strong German brand trust
• Institutional momentum: STACKIT is already selected by the Dutch government framework (SLM Rijk) and named on the European Commission’s sovereign cloud procurement framework

SKE support makes Mendix deployable on Europe’s sovereign infrastructure without changing how you build, deploy, or manage your applications. Whether sovereignty is a hard requirement or a competitive advantage, you now have the option.

For details on deploying Mendix on STACKIT Kubernetes Engine, see Mendix on Kubernetes.

Helm chart support in Mendix on Kubernetes

Deploy Mendix the way you deploy everything else

Mendix on Kubernetes now supports Helm chart-based installation and configuration as an alternative to the traditional mxpc-cli tool. If your IT operations teams already standardize on Helm and GitOps workflows, Mendix can now fit seamlessly into those existing pipelines instead of requiring a separate manual process.

The solution consists of two parts:

1. Helm charts: standard, declarative configuration for the Mendix Operator and namespace setup
2. mx-ops-cli: a local web UI tool that guides you through configuration via a wizard, then generates the Helm values file you deploy with standard helm commands.

Benefits over mxpc-cli

• No elevated permissions required: Helm-based installs can run from individual workstations, enabling GitOps and CI/CD practices that mxpc-cli’s permission model didn’t support
• Pipeline-native: integrates directly into existing CI/CD tools like ArgoCD, Flux, Tekton, or Jenkins, reducing manual intervention
• Familiar for operations teams: if you already know Helm, this approach is far more intuitive than manual CLI operations
• Repeatable and auditable: declarative configuration makes deployments consistent and traceable across environments

How it works

The mx-ops-cli wizard walks you through:

• General settings (environment type, namespace)
• Service accounts (optional custom Kubernetes service accounts)
• Database plans (PostgreSQL, SQL Server, dedicated JDBC)
• Storage plans (S3-compatible object storage, Azure Blob, MinIO)
• Ingress configuration (NGINX, Traefik, OpenShift routes)
• Registry settings (pull secrets for custom image registries)
• Proxy and custom TLS (for self-signed certificates)

Once configured, you download the generated YAML file and deploy with standard Helm commands:

This is an initial release. In future versions, we’re planning to publish the Helm charts and CRDs to a public OCI repository, so you can consume them directly without needing the mx-ops-cli tool for every download, making the workflow even more streamlined.

For the complete Helm chart setup guide, see Installing Components through the Helm Chart UI.

Siemens Xcelerator

Mendix inside Teamcenter Active Workspace (beta)

Teamcenter’s Active Workspace is excellent at managing PLM data, but many of the workflows surrounding that data — supplier qualification, quality non-conformance handling, sourcing preparation, guided engineering procedures — require integration with ERP systems, MES records, quality platforms, and approval workflow engines. Teamcenter provides native customization capabilities and connectors for many of these scenarios, but when something isn’t covered by available extensions, organizations need an alternative to heavy coding. That’s where Mendix helps, connecting systems rapidly and building custom UI and workflow logic without the traditional development overhead. Until now, though, that meant users had to leave Active Workspace to get there.

Mendix 11.12 introduces Mendix inside Teamcenter Active Workspace as a public beta, available on Mendix 11.12 and Teamcenter 2512. You can now embed a fully running Mendix application as a web component inside Active Workspace — not as a separate tab or popup, but as an integrated part of the AW interface. You configure which Teamcenter context attributes the Mendix app should receive, such as the selected item revision or BOM position, and from that point on, the platform propagates selection changes automatically as the user navigates Active Workspace. Once configured properly, authentication is handled by the platform, so users move seamlessly from Teamcenter into the embedded Mendix experience without being prompted to log in again. Teamcenter data is available inside the Mendix app via the TC Connector, and the Mendix app can read from and write to any other connected system at the same time.

Consider an engineer reviewing a part revision in Active Workspace. With an embedded Mendix app, they can see — in the same screen — whether the part has open non-conformances in the quality system, what the current stock level is in SAP, and an AI-generated recommendation on whether to proceed, request a deviation, or escalate. They can act on that recommendation directly: submit a deviation request, trigger a sourcing workflow, or notify a supplier — all powered by Mendix without leaving Teamcenter.

We strongly recommend involving Mendix Expert Services for your first implementation. The integration spans two platforms and requires careful coordination of authentication, network configuration, and application architecture — Expert Services teams have hands-on experience with early beta deployments. GA is targeted for Mendix 11.18 and Teamcenter 2612. More information on the public beta is available in our documentation.

Teamcenter Connector 2606 (Unified)

We’re excited to announce the release of Teamcenter Connector 2606 (available early July), a significant step forward in simplifying how developers build applications that integrate with Teamcenter.

This release brings two major improvements. First, Teamcenter Extension is now built directly into Teamcenter Connector, eliminating the need to install and manage multiple modules. With a single download and no version compatibility concerns, getting started is faster and easier than ever.
Second, Teamcenter Extension is now exposed as a custom document within Studio Pro. This new approach gives developers greater flexibility by allowing them to create multiple integration documents, connect to multiple Teamcenter environments during design time, and export or import integration configurations across projects.

Together, these enhancements provide a more streamlined, scalable, and intuitive development experience. By aligning more closely with the Studio Pro way of working, Teamcenter Connector 2606 makes it easier than ever to design, manage, and reuse Teamcenter integrations while focusing on what matters most—building great applications.

Ready to transform your next idea into an app?

• Kickstart your project by structuring your vision with Maia Plan.
• Download Mendix Studio Pro and use Maia Make to turn your plan into an app.
• Take a closer look at all the features, improvements, fixes, and more in the Mendix 11.12 release notes.
• Don’t want to miss future updates? Subscribe to our blog to catch the latest news