Skip to content

API Reference

Packages

litellm.litellm.ai/v1alpha1

Package v1alpha1 contains API Schema definitions for the litellm v1alpha1 API group.

Resource Types

ConnectionRef

Appears in: - ModelSpec

Field Description Default Validation
secretRef SecretRef
instanceRef InstanceRef

DatabaseSecretKeys

Appears in: - DatabaseSecretRef

Field Description Default Validation
hostSecret string
passwordSecret string
usernameSecret string
dbnameSecret string

DatabaseSecretRef

Appears in: - LiteLLMInstanceSpec

Field Description Default Validation
nameRef string
keys DatabaseSecretKeys

Gateway

Appears in: - LiteLLMInstanceSpec

Field Description Default Validation
enabled boolean
host string

Ingress

Appears in: - LiteLLMInstanceSpec

Field Description Default Validation
enabled boolean
host string

InitModelInstance

model instance used to create proxy server config map

Appears in: - LiteLLMInstanceSpec

Field Description Default Validation
modelName string
requiresAuth boolean
identifier string
modelCredentials ModelCredentialSecretRef
liteLLMParams LiteLLMParams
additionalProperties RawExtension

InstanceRef

Appears in: - ConnectionRef

Field Description Default Validation
namespace string
name string

LiteLLMInstance

LiteLLMInstance is the Schema for the litellminstances API.

Appears in: - LiteLLMInstanceList

Field Description Default Validation
apiVersion string litellm.litellm.ai/v1alpha1
kind string LiteLLMInstance
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 LiteLLMInstanceSpec
status LiteLLMInstanceStatus

LiteLLMInstanceList

LiteLLMInstanceList contains a list of LiteLLMInstance.

Field Description Default Validation
apiVersion string litellm.litellm.ai/v1alpha1
kind string LiteLLMInstanceList
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 LiteLLMInstance array

LiteLLMInstanceSpec

LiteLLMInstanceSpec defines the desired state of LiteLLMInstance.

Appears in: - LiteLLMInstance

Field Description Default Validation
image string ghcr.io/berriai/litellm-database:main-v1.74.9.rc.1
masterKey string
databaseSecretRef DatabaseSecretRef
redisSecretRef RedisSecretRef
ingress Ingress
gateway Gateway
replicas integer 1
models InitModelInstance array
extraEnvVars EnvVar array

LiteLLMInstanceStatus

LiteLLMInstanceStatus defines the observed state of LiteLLMInstance.

Appears in: - LiteLLMInstance

Field Description Default Validation
observedGeneration integer ObservedGeneration represents the .metadata.generation that the condition was set based upon
lastUpdated Time LastUpdated represents the last time the status was updated
configMapCreated boolean Resource creation status
secretCreated boolean
deploymentCreated boolean
serviceCreated boolean
ingressCreated boolean
conditions Condition array Conditions represent the latest available observations of a LiteLLM instance's state

LiteLLMParams

LiteLLMParams defines the LiteLLM parameters for a model.

Appears in: - InitModelInstance - ModelSpec - ModelStatus

Field Description Default Validation
inputCostPerToken string InputCostPerToken is the cost per input token
outputCostPerToken string OutputCostPerToken is the cost per output token
inputCostPerSecond string InputCostPerSecond is the cost per second for input
outputCostPerSecond string OutputCostPerSecond is the cost per second for output
inputCostPerPixel string InputCostPerPixel is the cost per pixel for input
outputCostPerPixel string OutputCostPerPixel is the cost per pixel for output
apiKey string APIKey is the API key for the model
apiBase string APIBase is the base URL for the API
apiVersion string APIVersion is the version of the API
vertexProject string VertexProject is the Google Cloud project for Vertex AI
vertexLocation string VertexLocation is the location for Vertex AI
vertexCredentials string VertexCredentials is the credentials for Vertex AI
regionName string RegionName is the region name for the service
awsAccessKeyId string AWSAccessKeyID is the AWS access key ID
awsSecretAccessKey string AWSSecretAccessKey is the AWS secret access key
awsRegionName string AWSRegionName is the AWS region name
watsonxRegionName string WatsonXRegionName is the WatsonX region name
customLLMProvider string CustomLLMProvider is the custom LLM provider
tpm integer TPM is tokens per minute
rpm integer RPM is requests per minute
timeout integer Timeout is the timeout in seconds
streamTimeout integer StreamTimeout is the stream timeout in seconds
maxRetries integer MaxRetries is the maximum number of retries
organization string Organization is the organization name
configurableClientsideAuthParams RawExtension ConfigurableClientsideAuthParams are configurable client-side auth parameters
litellmCredentialName string LiteLLMCredentialName is the LiteLLM credential name
litellmTraceId string LiteLLMTraceID is the LiteLLM trace ID
maxFileSizeMb integer MaxFileSizeMB is the maximum file size in MB
maxBudget string MaxBudget is the maximum budget
budgetDuration string BudgetDuration is the budget duration
useInPassThrough boolean UseInPassThrough indicates if to use in pass through
useLitellmProxy boolean UseLiteLLMProxy indicates if to use LiteLLM proxy
mergeReasoningContentInChoices boolean MergeReasoningContentInChoices indicates if to merge reasoning content in choices
modelInfo RawExtension ModelInfo contains additional model information
mockResponse string MockResponse is the mock response
autoRouterConfigPath string AutoRouterConfigPath is the auto router config path
autoRouterConfig string AutoRouterConfig is the auto router config
autoRouterDefaultModel string AutoRouterDefaultModel is the auto router default model
autoRouterEmbeddingModel string AutoRouterEmbeddingModel is the auto router embedding model
model string Model is the model name
additionalProp1 RawExtension AdditionalProps contains additional properties

Model

Model is the Schema for the models API.

Appears in: - ModelList

Field Description Default Validation
apiVersion string litellm.litellm.ai/v1alpha1
kind string Model
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 ModelSpec
status ModelStatus

ModelCredentialSecretKeys

Appears in: - ModelCredentialSecretRef

Field Description Default Validation
apiKey string
apiBase string
awsSecretAccessKey string
awsAccessKeyId string
vertexCredentials string
vertexProject string

ModelCredentialSecretRef

Appears in: - InitModelInstance

Field Description Default Validation
nameRef string
keys ModelCredentialSecretKeys

ModelInfo

ModelInfo defines the model information.

Appears in: - ModelSpec

Field Description Default Validation
id string ID is the model ID
dbModel boolean DBModel indicates if this is a database model
teamId string TeamID is the team ID
teamPublicModelName string TeamPublicModelName is the team public model name
additionalProp1 RawExtension AdditionalProps contains additional properties

ModelList

ModelList contains a list of Model.

Field Description Default Validation
apiVersion string litellm.litellm.ai/v1alpha1
kind string ModelList
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 Model array

ModelSpec

ModelSpec defines the desired state of Model.

Appears in: - Model

Field Description Default Validation
connectionRef ConnectionRef ConnectionRef is the connection reference
modelName string ModelName is the name of the model
litellmParams LiteLLMParams LiteLLMParams contains the LiteLLM parameters
modelInfo ModelInfo ModelInfo contains the model information
modelSecretRef SecretRef ModelSecretRef is the model secret reference

ModelStatus

ModelStatus defines the observed state of Model.

Appears in: - Model

Field Description Default Validation
observedGeneration integer ObservedGeneration represents the .metadata.generation that the condition was set based upon
lastUpdated Time LastUpdated represents the last time the status was updated
conditions Condition array Conditions represent the latest available observations of a LiteLLM instance's state
modelName string ModelName is the name of the model
litellmParams LiteLLMParams LiteLLMParams contains the LiteLLM parameters
modelId string ModelId contains the model uuid provided by litellm server

RedisSecretKeys

Appears in: - RedisSecretRef

Field Description Default Validation
hostSecret string
portSecret string
passwordSecret string

RedisSecretRef

Appears in: - LiteLLMInstanceSpec

Field Description Default Validation
nameRef string
keys RedisSecretKeys

SecretRef

Appears in: - ConnectionRef - ModelSpec

Field Description Default Validation
namespace string
secretName string