API Reference

Packages

• vllm.ai/v1alpha1

vllm.ai/v1alpha1

Package v1alpha1 contains API Schema definitions for the v1alpha1 API group

Resource Types

• IntelligentPool
• IntelligentPoolList
• IntelligentRoute
• IntelligentRouteList

Decision

Decision defines a routing decision based on rule combinations

Appears in:

• IntelligentRouteSpec

Field	Description	Default	Validation
name string	Name is the unique identifier for this decision		MaxLength: 100 MinLength: 1 Required: {}
priority integer	Priority defines the priority of this decision (higher values = higher priority) Used when strategy is "priority"	0	Maximum: 1000 Minimum: 0
description string	Description provides a human-readable description of this decision		MaxLength: 500
signals SignalCombination	Signals defines the signal combination logic		Required: {}
modelRefs ModelRef array	ModelRefs defines the model references for this decision (currently only one model is supported)		MaxItems: 1 MinItems: 1 Required: {}
plugins DecisionPlugin array	Plugins defines the plugins to apply for this decision		MaxItems: 10

DecisionPlugin

DecisionPlugin defines a plugin configuration for a decision

Appears in:

• Decision

Field	Description	Default	Validation
type string	Type is the plugin type (fast_response, hallucination, header_mutation, image_gen, memory, rag, request_params, response_jailbreak, router_replay, semantic-cache, system_prompt, tools)		Enum: [fast_response hallucination header_mutation image_gen memory rag request_params response_jailbreak router_replay semantic-cache system_prompt tools] Required: {}
configuration RawExtension	Configuration is the plugin-specific configuration as a raw JSON object		Schemaless: {}

DomainSignal

DomainSignal defines a domain category for classification

Appears in:

• Signals

Field	Description	Default	Validation
name string	Name is the unique identifier for this domain		MaxLength: 100 MinLength: 1 Required: {}
description string	Description provides a human-readable description of this domain		MaxLength: 500

EmbeddingSignal

EmbeddingSignal defines an embedding-based signal extraction rule

Appears in:

• Signals

Field	Description	Default	Validation
name string	Name is the unique identifier for this signal		MaxLength: 100 MinLength: 1 Required: {}
threshold float	Threshold is the similarity threshold for matching (0.0-1.0)		Maximum: 1 Minimum: 0 Required: {}
candidates string array	Candidates is the list of candidate phrases for semantic matching		MaxItems: 100 MinItems: 1 Required: {}
aggregationMethod string	AggregationMethod defines how to aggregate multiple candidate similarities	max	Enum: [mean max any]

IntelligentPool

IntelligentPool defines a pool of models with their configurations

Appears in:

• IntelligentPoolList

Field	Description	Default	Validation
apiVersion string	vllm.ai/v1alpha1
kind string	IntelligentPool
metadata ObjectMeta	Refer to Kubernetes API documentation for fields of metadata.
spec IntelligentPoolSpec
status IntelligentPoolStatus

IntelligentPoolList

IntelligentPoolList contains a list of IntelligentPool

Field	Description	Default	Validation
apiVersion string	vllm.ai/v1alpha1
kind string	IntelligentPoolList
metadata ListMeta	Refer to Kubernetes API documentation for fields of metadata.
items IntelligentPool array

IntelligentPoolSpec

IntelligentPoolSpec defines the desired state of IntelligentPool

Appears in:

• IntelligentPool

Field	Description	Default	Validation
defaultModel string	DefaultModel specifies the default model to use when no specific model is selected		MaxLength: 100 MinLength: 1 Required: {}
models ModelConfig array	Models defines the list of available models in this pool		MaxItems: 100 MinItems: 1 Required: {}

IntelligentPoolStatus

IntelligentPoolStatus defines the observed state of IntelligentPool

Appears in:

• IntelligentPool

Field	Description	Default	Validation
conditions Condition array	Conditions represent the latest available observations of the IntelligentPool's state
observedGeneration integer	ObservedGeneration reflects the generation of the most recently observed IntelligentPool
modelCount integer	ModelCount indicates the number of models in the pool

IntelligentRoute

IntelligentRoute defines intelligent routing rules and decisions

Appears in:

• IntelligentRouteList

Field	Description	Default	Validation
apiVersion string	vllm.ai/v1alpha1
kind string	IntelligentRoute
metadata ObjectMeta	Refer to Kubernetes API documentation for fields of metadata.
spec IntelligentRouteSpec
status IntelligentRouteStatus

IntelligentRouteList

IntelligentRouteList contains a list of IntelligentRoute

Field	Description	Default	Validation
apiVersion string	vllm.ai/v1alpha1
kind string	IntelligentRouteList
metadata ListMeta	Refer to Kubernetes API documentation for fields of metadata.
items IntelligentRoute array

IntelligentRouteSpec

IntelligentRouteSpec defines the desired state of IntelligentRoute

Appears in:

• IntelligentRoute

Field	Description	Default	Validation
signals Signals	Signals defines signal extraction rules for routing decisions
decisions Decision array	Decisions defines the routing decisions based on signal combinations		MaxItems: 100 MinItems: 1 Required: {}

IntelligentRouteStatus

IntelligentRouteStatus defines the observed state of IntelligentRoute

Appears in:

• IntelligentRoute

Field	Description	Default	Validation
conditions Condition array	Conditions represent the latest available observations of the IntelligentRoute's state
observedGeneration integer	ObservedGeneration reflects the generation of the most recently observed IntelligentRoute
statistics RouteStatistics	Statistics provides statistics about configured decisions and signals

KeywordSignal

KeywordSignal defines a keyword-based signal extraction rule

Appears in:

• Signals

Field	Description	Default	Validation
name string	Name is the unique identifier for this rule (also used as category name)		MaxLength: 100 MinLength: 1 Required: {}
operator string	Operator defines the logical operator for keywords (AND/OR)		Enum: [AND OR] Required: {}
keywords string array	Keywords is the list of keywords to match		MaxItems: 100 MinItems: 1 Required: {}
caseSensitive boolean	CaseSensitive specifies whether keyword matching is case-sensitive	false

LoRAConfig

LoRAConfig defines a LoRA adapter configuration

Appears in:

• ModelConfig

Field	Description	Default	Validation
name string	Name is the unique identifier for this LoRA adapter		MaxLength: 100 MinLength: 1 Required: {}
description string	Description provides a human-readable description of this LoRA adapter		MaxLength: 500

ModelConfig

ModelConfig defines the configuration for a single model

Appears in:

• IntelligentPoolSpec

Field	Description	Default	Validation
name string	Name is the unique identifier for this model		MaxLength: 100 MinLength: 1 Required: {}
reasoningFamily string	ReasoningFamily specifies the reasoning syntax family (e.g., "qwen3", "deepseek") Must be defined in the global static configuration's ReasoningFamilies		MaxLength: 50
pricing ModelPricing	Pricing defines the cost structure for this model
loras LoRAConfig array	LoRAs defines the list of LoRA adapters available for this model		MaxItems: 50

ModelPricing

ModelPricing defines the pricing structure for a model

Appears in:

• ModelConfig

Field	Description	Default	Validation
inputTokenPrice float	InputTokenPrice is the cost per input token		Minimum: 0
outputTokenPrice float	OutputTokenPrice is the cost per output token		Minimum: 0

ModelRef

ModelRef defines a model reference without score

Appears in:

• Decision

Field	Description	Default	Validation
model string	Model is the name of the model (must exist in IntelligentPool)		MaxLength: 100 MinLength: 1 Required: {}
loraName string	LoRAName is the name of the LoRA adapter to use (must exist in the model's LoRAs)		MaxLength: 100
useReasoning boolean	UseReasoning specifies whether to enable reasoning mode for this model	false
reasoningDescription string	ReasoningDescription provides context for when to use reasoning		MaxLength: 500
reasoningEffort string	ReasoningEffort defines the reasoning effort level (low/medium/high)		Enum: [low medium high]

RouteStatistics

RouteStatistics provides statistics about the IntelligentRoute configuration

Appears in:

• IntelligentRouteStatus

Field	Description	Default	Validation
decisions integer	Decisions indicates the number of decisions
keywords integer	Keywords indicates the number of keyword signals
embeddings integer	Embeddings indicates the number of embedding signals
domains integer	Domains indicates the number of domain signals

SignalCombination

SignalCombination defines how to combine multiple signals

Appears in:

• Decision

Field	Description	Default	Validation
operator string	Operator defines the logical operator for combining conditions (AND/OR)		Enum: [AND OR] Required: {}
conditions SignalCondition array	Conditions defines the list of signal conditions		MaxItems: 50 MinItems: 1 Required: {}

SignalCondition

SignalCondition defines a single signal condition

Appears in:

• SignalCombination

Field	Description	Default	Validation
type string	Type defines the type of signal (keyword/embedding/domain)		Enum: [keyword embedding domain] Required: {}
name string	Name is the name of the signal to reference		MaxLength: 100 MinLength: 1 Required: {}

Signals

Signals defines signal extraction rules

Appears in:

• IntelligentRouteSpec

Field	Description	Default	Validation
keywords KeywordSignal array	Keywords defines keyword-based signal extraction rules		MaxItems: 100
embeddings EmbeddingSignal array	Embeddings defines embedding-based signal extraction rules		MaxItems: 100
domains DomainSignal array	Domains defines MMLU domain categories for classification		MaxItems: 14