> ## 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.

# Deployment guidelines v5.12

> Component versions and deployment notes for FlowX.AI 5.12 (SaaS release).

> [!NOTE]
>
> After upgrading to a new platform version, always ensure that your installed component versions match the versions specified in the release notes. To verify this, navigate to **FlowX.AI Designer → Platform Status**.

## Upgrade overview

FlowX.AI 5.12 is a **SaaS-managed release**. Deployment-level changes, including infrastructure, Kafka topics, and service configuration, are handled automatically by FlowX. The information below is provided for transparency.

## Component versions

The versions below are the component versions that make up FlowX.AI 5.12.0. On SaaS these are managed by FlowX; the list is provided for transparency.

### Core components

| Component | 5.12.0 |
| - | - |
| **admin** | 11.69.2 |
| **advancing-controller** | 11.46.0 |
| **application-manager** | 11.68.3 |
| **audit-core** | 11.44.1 |
| **authorization-system** | 11.53.2 |
| **cms-core** | 11.61.1 |
| **data-search** | 11.44.1 |
| **data-sync** | 11.43.1 |
| **designer** | 11.126.1 |
| **document-plugin** | 11.68.1 |
| **email-gateway** | 11.58.1 |
| **events-gateway** | 11.49.2 |
| **file-gateway** | 11.55.1 |
| **integration-designer** | 11.90.2 |
| **license** | 11.48.1 |
| **nosql-db-runner** | 11.49.1 |
| **notification-plugin** | 11.61.2 |
| **organization-manager** | 11.53.1 |
| **process-engine** | 11.76.1 |
| **runtime-manager** | 11.68.3 |
| **scheduler-core** | 11.45.1 |
| **task-management-plugin** | 11.59.2 |
| **webhook-gateway** | 11.54.1 |

### AI base components

| Component | 5.12.0 |
| - | - |
| **embedder** | 11.5.0 |
| **knowledgebase-indexer-v2** | 11.5.0 |
| **knowledgebase-rag** | 11.5.0 |
| **doc-parser** | 11.5.0 |

### AI dev-time components

| Component | 5.12.0 |
| - | - |
| **ai-gateway** | 11.53.1 |
| **ai-analyst** | 11.5.0 |
| **ai-assistant** | 11.5.0 |
| **ai-designer** | 11.5.0 |
| **ai-developer** | 11.5.0 |
| **kb-enrichment** | 11.5.0 |
| **planner** | 11.5.0 |

### AI runtime components

| Component | 5.12.0 |
| - | - |
| **agent-builder** | 11.5.0 |
| **ai-architect** | 11.5.0 |
| **data-privacy** | 11.5.0 |
| **doc-converter** | 11.5.0 |
| **dynamic-classifier** | 11.5.0 |
| **evals-judge** | 11.5.0 |
| **speech-to-text** | 11.5.0 |
| **web-crawler** | 11.5.0 |

### AI extensions

| Component | 5.12.0 |
| - | - |
| **modpod** | 11.0.1 |
| **observatory-api** | 11.3.0 |
| **observatory-platform** | 11.3.0 |
| **observatory-web** | 11.3.0 |

### Embedded components

* **SpiceDB:** see the [5.9.0 deployment guidelines](https://docs.flowx.ai/release-notes/v5.x/v5.9.0-june-2026/deployment-guidelines-v5.9.md) for details.
* **Qdrant** (vector database): bundled in the AI Platform Helm chart; its version is managed by the chart and not pinned here. See the [Qdrant setup guide](https://docs.flowx.ai/5.9/setup-guides/qdrant.md) for deployment and configuration.

### Renderers

The web SDKs ship at the **designer** version (`11.126.1` for FlowX.AI 5.12.0). The native iOS and Android renderers version independently.

| Component | 5.12.0 |
| - | - |
| **@flowx/angular-sdk** | 11.126.1 |
| **@flowx/angular-theme** | 11.126.1 |
| **@flowx/angular-ui-toolkit** | 11.126.1 |
| **@flowx/react-sdk** | 11.126.1 |
| **@flowx/react-theme** | 11.126.1 |
| **@flowx/react-ui-toolkit** | 11.126.1 |
| **@flowx/react-native-sdk** | 11.126.1 |
| **@flowx/react-native-theme** | 11.126.1 |
| **@flowx/react-native-ui-toolkit** | 11.126.1 |
| **@flowx/core-sdk** | 11.126.1 |
| **@flowx/core-theme** | 11.126.1 |
| **react-saas** | 11.126.1 |
| **iOS renderer** | TBD |
| **Android renderer** | TBD |

The **react-saas** container app tracks the designer version.

## Configuration changes

> [!NOTE]
>
> On SaaS, required configuration changes are applied automatically by FlowX. The changes below are listed for transparency. Self-hosted deployments will receive them, re-verified, with the next LTS release family.

### New Kafka topics

These topics back new 5.12 capabilities. On SaaS they are created and managed automatically.

**Event-triggered workflows** (email and webhook triggers can start an integration workflow):

| Topic | Producer → Consumer |
| - | - |
| `ai.flowx.build.trigger-subscription-change.in.v1` | admin, integration-designer → application-manager |
| `ai.flowx.core.trigger.start-for-event.workflow.v1` | email-gateway, webhook-gateway → integration-designer |

**Deep Research node** (the node hands research jobs to the AI Platform over Kafka):

| Topic | Producer → Consumer |
| - | - |
| `ai.flowx.ai-platform.deep-research.job.request.v1` | integration-designer → agent-builder |
| `ai.flowx.ai-platform.deep-research.job.response.v1` | agent-builder → integration-designer |
| `ai.flowx.ai-platform.deep-research.job.dlq.v1` | dead-letter queue for failed research jobs |

**Bulk process-instance migration:**

| Topic | Producer → Consumer |
| - | - |
| `ai.flowx.core.trigger.instance-migration.batch.v1` | process-engine → process-engine |

**Notifications:**

| Topic | Producer → Consumer |
| - | - |
| `ai.flowx.eventsgateway.user.notifications.v1` | application-manager, process-engine → events-gateway |

**Document classification:**

| Topic | Producer → Consumer |
| - | - |
| `ai.flowx.dynamic-classifier.job.dlq.v1` | dynamic-classifier → dead-letter sink (not consumed) |

### New service dependency: webhook-gateway and MongoDB

The webhook gateway now reads the runtime registry that backs event-triggered workflows, so it depends on the runtime MongoDB database in addition to its existing relational database. On SaaS this connection is provisioned and managed by FlowX.

### New required secret: personal-data encryption master key

The Mask Personal Data and Restore Personal Data nodes encrypt the placeholder map they keep during a workflow run. The key material comes from a new integration-designer variable with **no default**: without it, the mask and restore nodes fail rather than encrypt personal data with a published value. On SaaS this secret is provisioned and managed by FlowX.

| Environment Variable | Description | Default Value | Component |
| - | - | - | - |
| `FLOWX_PII_ENCRYPTION_MASTER_KEY` | Master secret the mask and restore nodes derive their encryption key from. Any non-empty string; use a strong random value and store it as a secret. Must be identical across all pods and stable over time: changing it invalidates every in-flight placeholder map. | none (required) | integration-designer |
| `FLOWX_PII_ENCRYPTION_KEY_ID` | Label recorded on new placeholder-map entries so a future migration can support key rotation. Only the configured key is held today. | `k1` | integration-designer |
| `FLOWX_PII_MAP_TTL` | How long a masked value stays restorable. Bounds exposure when a run ends without reaching the instance-end cleanup hooks; enforced by a MongoDB TTL index. | `PT24H` | integration-designer |

### Native PostgreSQL partitioning for the engine

On PostgreSQL, the engine converts its runtime tables to native `RANGE(partition_id)` partitioning the first time it starts on the new version, and archiving then detaches a partition instead of copying rows out of it. Deployments that never enabled partitioning are skipped entirely, and Oracle deployments are unaffected.

Unlike most 5.12 changes, this one also reaches self-hosted 5.9.x in the 5.9.5 patch rather than waiting for the next LTS release family.

| Environment Variable | Description | Default Value | Component |
| - | - | - | - |
| `FLOWX_DATA_PARTITIONING_ARCHIVING_POSTGRESNATIVE` | Whether archiving detaches partitions with `DETACH CONCURRENTLY` on natively partitioned tables. Setting it to `false` returns archiving to the previous batched insert-and-delete, but does not undo the conversion: the tables stay natively partitioned. | `true` | process-engine |

For the conversion cost, table locking and the one-way caveat, see [Process instance data archiving](https://docs.flowx.ai/5.9/setup-guides/flowx-engine-setup-guide/process-instance-data-archiving.md#native-postgresql-partitioning).

> [!NOTE]
>
> For previous FlowX.AI releases, please refer to the [Release Notes overview](https://docs.flowx.ai/release-notes/overview.md).


## Related topics

- [FlowX.AI 5.12.0 Release Notes](https://docs.flowx.ai/release-notes/v5.x/v5.12.0-september-2026/v5.12.0-september-2026.md)
- [Deployment guidelines v5.1.2](https://docs.flowx.ai/release-notes/v5.x/v5.1.x-lts/deployment-guidelines-v5.1.2.md)
- [Deployment guidelines v5.1.12](https://docs.flowx.ai/release-notes/v5.x/v5.1.x-lts/deployment-guidelines-v5.1.12.md)
- [Deployment guidelines v5.0.0](https://docs.flowx.ai/release-notes/v5.x/v5.0.0-july-2025/deployment-guidelines-v5.0.md)
- [Deployment guidelines v5.1.0](https://docs.flowx.ai/release-notes/v5.x/v5.1.x-lts/v5.1.0-october-2025/deployment-guidelines-v5.1.md)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.