SUSE AI Factory
Configuring artificial intelligence (AI) applications in the Kubernetes environment involves using Helm charts and custom values files. While applications ship with default configurations, you can modify application behavior by applying a custom values file with scenario-specific values. This approach ensures consistent, reproducible and easily manageable application deployment.
This guide provides a list of AI applications installable by SUSE AI Factory, along with their descriptions and configuration override examples.
Publication Date: 2026-11-12
- 1 Kubeflow
- 1.1 What components does Kubeflow include?
- 1.2 What are Kubeflow requirements?
- 1.3 How do I get the values list for Kubeflow Helm chart?
- 1.4 How do I install Kubeflow?
- 1.5 How do I install Kubeflow in air-gapped environments?
- 1.6 How do I access Kubeflow?
- 1.7 How do I configure Kubeflow for different scenarios?
- 1.8 How do I harden Kubeflow for production use?
- 1.9 How do I manage user profiles and namespaces?
- 1.10 How do I upgrade Kubeflow?
- 1.11 How do I remove Kubeflow?
- 1.12 Troubleshooting
- 2 LiteLLM
- 3 Milvus
- 4 Ollama
- 5 Open WebUI
- 6 PyTorch
- 7 Qdrant
- 8 vLLM
- 9 MLflow
List of Figures
List of Examples
- 4.1 Enabling or disabling the GPU section
- 4.2 Basic custom values file with GPU and two models pulled at startup
- 4.3 Basic custom values file with Ingress and no GPU
- 5.1 Open WebUI custom values file with Ollama included
- 5.2 Open WebUI custom values file with Ollama installed separately
- 5.3 Open WebUI custom values file with pipelines enabled
- 5.4 Open WebUI custom values file with a connection to vLLM
- 5.5 Stand-alone deployment of open-webui-pipelines
- 6.1 Basic custom values file with GPU enabled
- 6.2 ConfigMap-based upload
- 6.3 Host-folder with files baked into the chart
- 6.4 Git repository clone: public with no authentication
- 6.5 Git repository clone: private with authentication
- 7.1 Basic custom values file when the cluster has no default storage class set.
- 7.2 An example where Qdrant uses GPU capabilities.
- 8.1 Minimal configuration
- 8.2 Validating the installation
- 8.3 Basic configuration
- 8.4 Loading prefetched models from persistent storage
- 8.5 Configuration with multiple models
- 8.6 CPU offloading
- 8.7 Shared remote KV cache storage with LMCache