> For the complete documentation index, see [llms.txt](https://doc.ozforensics.com/oz-knowledge/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://doc.ozforensics.com/oz-knowledge/guides/administrator-guide/updates/how-to-update-models.md).

# How to Update Models

Our product in the Full API version is delivered in two ways:

* via Helm charts,
* via Docker Compose.

For Instant API, models are not updated. Instead, a new Instant API image is released and delivered.

## Kubernetes installations (delivery via Helm charts)

1. Support engineers notify the client of a new model version through the relevant communication channel.
2. The client confirms they are ready to update the model version and specifies a suitable time for the work. The client coordinates with their information security team and management on their own.
3. An engineer prepares the update for the corresponding client environment (DEV or PROD) via the `Oz-Bio-Admin` service.

Our `oz-k8s-bio-updater` service on the client side polls our `https://bio-admin.infra.ozforensics.com` service on its own. For this, the client must allow access to the following resources:

* `https://*.infra.ozforensics.ai/`
* `https://*.infra.ozforensics.com`
* `https://*.s3.amazonaws.com/`
* `https://*.s3-accelerate.amazonaws.com/`

If the oz-k8s-bio-updater service detects an update, it downloads and unpacks the full set of models to its PVC. The client can monitor this process via the `oz-k8s-bio-updater` StatefulSet log.

The update takes effect only after all `oz-k8s-bio` pods are restarted manually. For example, via `kubectl`:

```bash
kubectl -n oz-namespace rollout restart deployment oz-k8s-bio
```

or via the ArgoCD GitOps tool:

```bash
argocd app actions run oz-k8s restart \
  --kind Deployment \
  --resource-name oz-k8s-bio \
  --namespace oz-namespace
```

To check that the update was successful, run curl on any `oz-k8s-bio` pod of the Kubernetes cluster:

```bash
curl localhost:8501/v1/models/inquisitor_c
```

Example response showing the `inquisitor_c` model and its current version:

<figure><img src="https://2532558063-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F5g6dgsxRbyrCvB0uAf8f%2Fuploads%2F22t6bOvqLogwLwnYjK4i%2Fk8s.png?alt=media&amp;token=e4fbb438-0876-409a-ba0e-68d41176f38d" alt=""><figcaption></figcaption></figure>

The update is considered successful only after the client confirms that the model version has been updated.

## Docker installations (delivery via Docker Compose)

1. Support engineers notify the client of a new model version through the relevant communication channel.
2. The client confirms they are ready to update the model version and specifies a suitable time for the work. The client coordinates with their information security team and management on their own.
3. An engineer prepares the update for the corresponding client environment (DEV or PROD) via the Oz-Bio-Admin service.

The `oz-bio-updater` container on the client side checks our `https://bio-admin.infra.ozforensics.com` service for new model updates on its own. For this, the client must allow access to the following resources:

* `https://*.infra.ozforensics.ai/`
* `https://*.infra.ozforensics.com`
* `https://*.s3.amazonaws.com/`
* `https://*.s3-accelerate.amazonaws.com/`

If the `oz-bio-updater` container detects an update, it downloads and unpacks the full set of models. The client can monitor this process via the container log.

Example command:

```bash
sudo docker logs -f oz-bio-updater
```

To apply the update, take the `oz-bio` containers out of load balancing one by one manually and restart them. Example command for an installation without a load balancer:

```bash
sudo docker compose -f docker-compose-bio.yml down && sudo docker compose -f docker-compose-bio.yml up -d
```

You can monitor the container restart via its log.

To check the current model version, run curl inside the oz-bio container:

```bash
sudo docker exec -it oz-bio curl -s localhost:8501/v1/models/inquisitor_c
```

Example of a successful response:

<figure><img src="https://2532558063-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F5g6dgsxRbyrCvB0uAf8f%2Fuploads%2Fc6vrWKAPxaUFxnmBAdGb%2Fdocker.png?alt=media&amp;token=65f31afa-7f98-4413-9c55-4eb22d51ef36" alt=""><figcaption></figcaption></figure>

The update is considered successful only after the client confirms that the model version has been updated.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://doc.ozforensics.com/oz-knowledge/guides/administrator-guide/updates/how-to-update-models.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
