> ## Documentation Index
> Fetch the complete documentation index at: https://docs.flowx.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Third-party components

> FlowX.AI uses a number of third-party software components

## Overview

FlowX.AI relies on various third-party components to deliver its functionality. These components fall into two categories:

* **Embedded components**: Third-party software packaged and managed by FlowX.AI
* **Client-managed components**: Third-party software that you deploy and maintain

<Info>
  For optimal performance and reliability, our internal QA process validates new releases using specific versions. While you can explore alternative versions that suit your requirements, we recommend referring to the compatibility matrix for guidance.
</Info>

<Note>
  Licensing of third-party components is governed by each vendor and may change over time. Consult each vendor's official licensing documentation for current terms.
</Note>

## Embedded components

<Note>
  Embedded components are shipped and configured as part of the FlowX.AI platform. FlowX.AI manages their versioning, patching, security updates, and provides L4 support coverage. They are deployed alongside the platform using the standard FlowX.AI deployment process.
</Note>

### Authorization

<CardGroup cols={2}>
  <Card title="SpiceDB" icon="shield" href="https://authzed.com/spicedb">
    An open-source permissions database inspired by Google's Zanzibar paper. Powers FlowX.AI's multi-tenant authorization system with fine-grained access control and relationship-based permissions.

    **Key capabilities**: Google Zanzibar-inspired authorization, ReBAC, global consistency, reverse lookup, sub-millisecond response times
  </Card>

  <Card title="SpiceDB Operator" icon="gears" href="https://github.com/authzed/spicedb-operator">
    A Kubernetes operator that automates the deployment, configuration, and lifecycle management of SpiceDB instances. Handles upgrades, scaling, and configuration changes for consistent and reliable SpiceDB deployments.
  </Card>

  <Card title="Qdrant" icon="cube" href="https://qdrant.tech/documentation/">
    **Supported Version**: 1.16.3

    Vector search engine used by the AI Platform RAG path (knowledgebase-rag, embedder, knowledgebase-indexer-v2) and by AI agents for semantic retrieval. Required for AI Platform deployments starting with 5.9.0.
  </Card>
</CardGroup>

## Client-managed components

<Info>
  Client-managed components are deployed and maintained by your team. FlowX.AI validates compatibility with supported versions listed in the [compatibility matrix](#compatibility-matrix) but does not provide direct support for these components.
</Info>

### Core infrastructure

<Tabs>
  <Tab title="Identity & access">
    <Card title="Keycloak" icon="key" href="https://www.keycloak.org/documentation">
      Open-source identity and access management solution enabling single sign-on for modern applications.
    </Card>
  </Tab>

  <Tab title="Messaging">
    <CardGroup cols={2}>
      <Card title="Apache Kafka" icon="message-dots" href="https://kafka.apache.org/documentation/">
        Distributed event streaming platform for high-volume, real-time data feeds with fault tolerance and low latency.
      </Card>

      <Card title="Elasticsearch Sink Connector" icon="arrows-turn-to-dots" href="https://docs.confluent.io/kafka-connectors/elasticsearch/current/overview.html">
        Kafka Connect sink plugin that streams process-instance events into Elasticsearch indices. Required only if you enable indexing via Kafka in the process engine. Other sink connector implementations also work, as long as they are compatible with both the deployed Kafka and Elasticsearch versions. See [Configuring Elasticsearch indexing](/5.9/setup-guides/flowx-engine-setup-guide/configuring-elasticsearch-indexing/elasticsearch-indexing).
      </Card>
    </CardGroup>
  </Tab>

  <Tab title="Databases">
    <CardGroup cols={2}>
      <Card title="PostgreSQL" icon="database" href="https://www.postgresql.org/docs/">
        Free and open-source relational database emphasizing extensibility and SQL compliance.
      </Card>

      <Card title="MongoDB" icon="database" href="https://www.mongodb.com/docs/">
        Document-oriented NoSQL database for business process data and configuration storage.
      </Card>
    </CardGroup>
  </Tab>

  <Tab title="Caching & storage">
    <CardGroup cols={2}>
      <Card title="Redis" icon="fire" href="https://redis.io/docs/">
        Fast in-memory key-value store for caching and pub/sub messaging with sub-millisecond response times.
      </Card>
    </CardGroup>
  </Tab>
</Tabs>

### Infrastructure & monitoring

<AccordionGroup>
  <Accordion title="Web server & load balancing">
    **NGINX Ingress Controller**

    Kubernetes ingress controller for routing web traffic from SPAs to backend services and the FlowX.AI Engine.

    <Note>
      Starting with 5.9.x, the FlowX.AI chart no longer assumes ingress-nginx as the bundled controller. Kubernetes Ingress remains fully supported with any compliant controller (Traefik, HAProxy, Istio, cloud-provider controllers), and Gateway API HTTPRoutes are supported as an alternative. See [Ingress routing and CORS changes](/5.9/migrating-from-5.1-lts/ingress-routing).
    </Note>

    <Card title="ingress-nginx documentation" href="https://kubernetes.github.io/ingress-nginx/" icon="external-link" />
  </Accordion>

  <Accordion title="Search & analytics">
    **Elasticsearch (EFK Stack)**

    Distributed search and analytics engine for centralized data storage and business process transaction searching.

    <CardGroup cols={2}>
      <Card title="Elastic Stack" href="https://www.elastic.co/elastic-stack/" icon="external-link" />

      <Card title="Fluentd" href="https://docs.fluentd.org/" icon="external-link" />
    </CardGroup>
  </Accordion>

  <Accordion title="Business intelligence">
    **Apache Superset**

    Business intelligence web application for data exploration and visualization, from simple charts to detailed dashboards.

    <Card title="Superset documentation" href="https://superset.apache.org/docs/intro" icon="external-link" />
  </Accordion>
</AccordionGroup>

### Commercial components

<Card title="Oracle Database" icon="database" href="https://www.oracle.com/database/technologies/">
  Enterprise relational database management system for mission-critical workloads.

  <Note>
    Oracle Database requires commercial licensing. Contact Oracle for licensing information.
  </Note>
</Card>

## Compatibility matrix

Within each version line listed below, versions that have reached upstream end of life are not supported, even though their major/minor is listed.

| FlowX.AI Version | 3rd Party Dependency              | Supported Versions |
| ---------------- | --------------------------------- | ------------------ |
| 5.9.x            | Keycloak                          | 26                 |
| 5.9.x            | Kafka                             | 3.9, 4.0, 4.1, 4.2 |
| 5.9.x            | PostgreSQL                        | 16, 17, 18         |
| 5.9.x            | Oracle Database                   | 21c, 23ai          |
| 5.9.x            | MongoDB                           | 7, 8               |
| 5.9.x            | Redis                             | 7, 8               |
| 5.9.x            | Elasticsearch                     | 8, 9               |
| 5.9.x            | Elasticsearch Sink Connector (\*) | 15                 |
| 5.9.x            | Angular (Web SDK)                 | 20.3               |
| 5.9.x            | React (Web SDK)                   | 19                 |

(\*) Required only if using indexing via Kafka. Multiple sink connector implementations also work, as long as they are compatible with both the deployed Kafka and Elasticsearch version.

<Info>
  The table lists version lines, not individual versions. A version is supported for as long as its vendor still supports it, so check the version you plan to deploy against that vendor's lifecycle page before committing to it. When a version reaches upstream end of life it is no longer supported; upgrade that component to keep your deployment on a supported configuration. Versions can also be added during the family's lifetime as FlowX validates them.
</Info>

## Deprecation notice

<Warning>
  The following versions are no longer supported. Versions are dropped from the supported list as support upstream is no longer provided.

  * Keycloak versions older than 26
  * Kafka versions older than 3.9 (3.8 dropped in 5.9.0)
</Warning>

## Support information

<Info>
  If you encounter compatibility issues with any of the supported versions listed above, please [open a support ticket](https://support.flowx.ai/) and our team will address any identified bugs through our standard support process.
</Info>

## Related resources

<CardGroup cols={2}>
  <Card title="Event-driven Architecture" href="../platform-overview/frameworks-and-standards/event-driven-architecture-frameworks/intro-to-kafka-concepts" icon="book">
    Learn about Kafka concepts and implementation
  </Card>

  <Card title="NGINX Configuration" href="../platform-overview/frameworks-and-standards/event-driven-architecture-frameworks/intro-to-nginx" icon="book">
    Understanding NGINX in FlowX.AI architecture
  </Card>

  <Card title="Data Architecture" href="../platform-overview/data-architecture" icon="database">
    Which data stores FlowX.AI uses and what each stores
  </Card>
</CardGroup>


## Related topics

- [Deployment guidelines v4.7.5](/release-notes/v4.x/v4.7.x-lts/v4.7.5-june-2025/deployment-guidelines-v4.7.5.md)
- [Deployment guidelines v4.7.6](/release-notes/v4.x/v4.7.x-lts/v4.7.6-july-2025/deployment-guidelines-v4.7.6.md)
- [Deployment guidelines v4.7.9](/release-notes/v4.x/v4.7.x-lts/v4.7.9-october-2025/deployment-guidelines-v4.7.9.md)
- [Deployment guidelines v4.7.7](/release-notes/v4.x/v4.7.x-lts/v4.7.7-august-2025/deployment-guidelines-v4.7.7.md)
- [Deployment guidelines v4.7.8](/release-notes/v4.x/v4.7.x-lts/v4.7.8-september-2025/deployment-guidelines-v4.7.8.md)