api-overview.md

API Reference

Packages

• llamastack.io/v1alpha1

llamastack.io/v1alpha1

Package v1alpha1 contains API Schema definitions for the v1alpha1 API group

Resource Types

• LlamaStackDistribution
• LlamaStackDistributionList

AllowedFromSpec

AllowedFromSpec defines namespace-based access controls for NetworkPolicies.

Appears in:

• NetworkSpec

Field	Description	Default	Validation
namespaces string array	Namespaces is an explicit list of namespace names allowed to access the service. Use "*" to allow all namespaces.
labels string array	Labels is a list of namespace label keys that are allowed to access the service. A namespace matching any of these labels will be granted access (OR semantics). Example: ["myproject/lls-allowed", "team/authorized"]

AutoscalingSpec

AutoscalingSpec configures HorizontalPodAutoscaler targets.

Appears in:

• ServerSpec

Field	Description	Default	Validation
minReplicas integer	MinReplicas is the lower bound replica count maintained by the HPA
maxReplicas integer	MaxReplicas is the upper bound replica count maintained by the HPA
targetCPUUtilizationPercentage integer	TargetCPUUtilizationPercentage configures CPU based scaling
targetMemoryUtilizationPercentage integer	TargetMemoryUtilizationPercentage configures memory based scaling

CABundleConfig

CABundleConfig defines the CA bundle configuration for custom certificates

Appears in:

• TLSConfig

Field	Description	Default	Validation
configMapName string	ConfigMapName is the name of the ConfigMap containing CA bundle certificates
configMapNamespace string	ConfigMapNamespace is the namespace of the ConfigMap (defaults to the same namespace as the CR)
configMapKeys string array	ConfigMapKeys specifies multiple keys within the ConfigMap containing CA bundle data All certificates from these keys will be concatenated into a single CA bundle file If not specified, defaults to [DefaultCABundleKey]		MaxItems: 50

ContainerSpec

ContainerSpec defines the llama-stack server container configuration.

Appears in:

• ServerSpec

Field	Description	Default	Validation
name string		llama-stack
port integer
resources ResourceRequirements
env EnvVar array
command string array
args string array

DistributionConfig

DistributionConfig represents the configuration information from the providers endpoint.

Appears in:

• LlamaStackDistributionStatus

Field	Description	Default	Validation
activeDistribution string	ActiveDistribution shows which distribution is currently being used
providers ProviderInfo array
availableDistributions object (keys:string, values:string)	AvailableDistributions lists all available distributions and their images

DistributionPhase

Underlying type: string

LlamaStackDistributionPhase represents the current phase of the LlamaStackDistribution

Validation:

• Enum: [Pending Initializing Ready Failed Terminating]

Appears in:

• LlamaStackDistributionStatus

Field	Description
Pending	LlamaStackDistributionPhasePending indicates that the distribution is pending initialization
Initializing	LlamaStackDistributionPhaseInitializing indicates that the distribution is being initialized
Ready	LlamaStackDistributionPhaseReady indicates that the distribution is ready to use
Failed	LlamaStackDistributionPhaseFailed indicates that the distribution has failed
Terminating	LlamaStackDistributionPhaseTerminating indicates that the distribution is being terminated

DistributionType

DistributionType defines the distribution configuration for llama-stack.

Appears in:

• ServerSpec

Field	Description	Default	Validation
name string	Name is the distribution name that maps to supported distributions.
image string	Image is the direct container image reference to use

LlamaStackDistribution

Appears in:

• LlamaStackDistributionList

Field	Description	Default	Validation
apiVersion string	llamastack.io/v1alpha1
kind string	LlamaStackDistribution
kind string	Kind is a string value representing the REST resource this object represents. Servers may infer this from the endpoint the client submits requests to. Cannot be updated. In CamelCase. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds
apiVersion string	APIVersion defines the versioned schema of this representation of an object. Servers should convert recognized schemas to the latest internal value, and may reject unrecognized values. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources
metadata ObjectMeta	Refer to Kubernetes API documentation for fields of metadata.
spec LlamaStackDistributionSpec
status LlamaStackDistributionStatus

LlamaStackDistributionList

LlamaStackDistributionList contains a list of LlamaStackDistribution.

Field	Description	Default	Validation
apiVersion string	llamastack.io/v1alpha1
kind string	LlamaStackDistributionList
kind string	Kind is a string value representing the REST resource this object represents. Servers may infer this from the endpoint the client submits requests to. Cannot be updated. In CamelCase. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds
apiVersion string	APIVersion defines the versioned schema of this representation of an object. Servers should convert recognized schemas to the latest internal value, and may reject unrecognized values. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources
metadata ListMeta	Refer to Kubernetes API documentation for fields of metadata.
items LlamaStackDistribution array

LlamaStackDistributionSpec

LlamaStackDistributionSpec defines the desired state of LlamaStackDistribution.

Appears in:

• LlamaStackDistribution

Field	Description	Default	Validation
replicas integer		1
server ServerSpec
network NetworkSpec	Network defines network access controls for the LlamaStack service

LlamaStackDistributionStatus

LlamaStackDistributionStatus defines the observed state of LlamaStackDistribution.

Appears in:

• LlamaStackDistribution

Field	Description	Default	Validation
phase DistributionPhase	Phase represents the current phase of the distribution		Enum: [Pending Initializing Ready Failed Terminating]
version VersionInfo	Version contains version information for both operator and deployment
distributionConfig DistributionConfig	DistributionConfig contains the configuration information from the providers endpoint
conditions Condition array	Conditions represent the latest available observations of the distribution's current state
availableReplicas integer	AvailableReplicas is the number of available replicas
serviceURL string	ServiceURL is the internal Kubernetes service URL where the distribution is exposed
routeURL string	RouteURL is the external URL where the distribution is exposed (when exposeRoute is true). nil when external access is not configured, empty string when Ingress exists but URL not ready.

NetworkSpec

NetworkSpec defines network access controls for the LlamaStack service.

Appears in:

• LlamaStackDistributionSpec

Field	Description	Default	Validation
exposeRoute boolean	ExposeRoute when true, creates an Ingress for external access. Default is false (internal access only).	false
allowedFrom AllowedFromSpec	AllowedFrom defines which namespaces are allowed to access the LlamaStack service. By default, only the LLSD namespace and the operator namespace are allowed.

PodDisruptionBudgetSpec

PodDisruptionBudgetSpec defines voluntary disruption controls.

Appears in:

• ServerSpec

Field	Description	Default	Validation
minAvailable IntOrString	MinAvailable is the minimum number of pods that must remain available
maxUnavailable IntOrString	MaxUnavailable is the maximum number of pods that can be disrupted simultaneously

PodOverrides

PodOverrides allows advanced pod-level customization.

Appears in:

• ServerSpec

Field	Description	Default	Validation
serviceAccountName string	ServiceAccountName allows users to specify their own ServiceAccount If not specified, the operator will use the default ServiceAccount
terminationGracePeriodSeconds integer	TerminationGracePeriodSeconds is the time allowed for graceful pod shutdown. If not specified, Kubernetes defaults to 30 seconds.
volumes Volume array
volumeMounts VolumeMount array

ProviderHealthStatus

HealthStatus represents the health status of a provider

Appears in:

• ProviderInfo

Field	Description	Default	Validation
status string
message string

ProviderInfo

ProviderInfo represents a single provider from the providers endpoint.

Appears in:

• DistributionConfig

Field	Description	Default	Validation
api string
provider_id string
provider_type string
config JSON
health ProviderHealthStatus

ServerSpec

ServerSpec defines the desired state of llama server.

Appears in:

• LlamaStackDistributionSpec

Field	Description	Default	Validation
distribution DistributionType
containerSpec ContainerSpec
workers integer	Workers configures the number of uvicorn worker processes to run. When set, the operator will launch llama-stack using uvicorn with the specified worker count. Ref: https://fastapi.tiangolo.com/deployment/server-workers/ CPU requests are set to the number of workers when set, otherwise 1 full core		Minimum: 1
podOverrides PodOverrides
podDisruptionBudget PodDisruptionBudgetSpec	PodDisruptionBudget controls voluntary disruption tolerance for the server pods
topologySpreadConstraints TopologySpreadConstraint array	TopologySpreadConstraints defines fine-grained spreading rules
autoscaling AutoscalingSpec	Autoscaling configures HorizontalPodAutoscaler for the server pods
storage StorageSpec	Storage defines the persistent storage configuration
userConfig UserConfigSpec	UserConfig defines the user configuration for the llama-stack server
tlsConfig TLSConfig	TLSConfig defines the TLS configuration for the llama-stack server

StorageSpec

StorageSpec defines the persistent storage configuration

Appears in:

• ServerSpec

Field	Description	Default	Validation
size Quantity	Size is the size of the persistent volume claim created for holding persistent data of the llama-stack server
mountPath string	MountPath is the path where the storage will be mounted in the container

TLSConfig

TLSConfig defines the TLS configuration for the llama-stack server

Appears in:

• ServerSpec

Field	Description	Default	Validation
caBundle CABundleConfig	CABundle defines the CA bundle configuration for custom certificates

UserConfigSpec

Appears in:

• ServerSpec

Field	Description	Default	Validation
configMapName string	ConfigMapName is the name of the ConfigMap containing user configuration
configMapNamespace string	ConfigMapNamespace is the namespace of the ConfigMap (defaults to the same namespace as the CR)

VersionInfo

VersionInfo contains version-related information

Appears in:

• LlamaStackDistributionStatus

Field	Description	Default	Validation
operatorVersion string	OperatorVersion is the version of the operator managing this distribution
llamaStackServerVersion string	LlamaStackServerVersion is the version of the LlamaStack server
lastUpdated Time	LastUpdated represents when the version information was last updated