Class OpenAIEmbedCapabilityImpl

Description

Provider capability implementation for OpenAIEmbedCapabilityImpl.

Implements

EmbedCapability<ClientEmbeddingRequest, NormalizedEmbedding[]>

Index

Constructors

Methods

Constructors

constructor

new OpenAIEmbedCapabilityImpl(
provider: BaseProvider,
client: OpenAI,
): OpenAIEmbedCapabilityImpl
Constructs a new OpenAI embedding capability.
Parameters
- provider: BaseProvider
  Owning provider instance
- client: OpenAI
  Initialized OpenAI SDK client
Returns OpenAIEmbedCapabilityImpl
- Defined in src/providers/openai/capabilities/OpenAIEmbedCapabilityImpl.ts:44

Methods

embed

embed(
    request: AIRequest<ClientEmbeddingRequest>,
    _executionContext?: MultiModalExecutionContext,
    signal?: AbortSignal,
): Promise<AIResponse<NormalizedEmbedding[]>>
Generates embeddings for one or more input strings.

Flow:
- Validate input
- Initialize capability execution context
- Normalize input to array for batch processing
- Call OpenAI embeddings API
- Extract numeric vectors, preserving order
- Return normalized AIResponse with metadata
Parameters
- request: AIRequest<ClientEmbeddingRequest>
  Unified embedding request
- Optional_executionContext: MultiModalExecutionContext
  Optional execution context
- Optionalsignal: AbortSignal
  Optional abort signal
Returns Promise<AIResponse<NormalizedEmbedding[]>>
AIResponse containing a single vector or array of vectors

Throws
Error if input is invalid or API returns no embeddings
Implementation of EmbedCapability.embed
- Defined in src/providers/openai/capabilities/OpenAIEmbedCapabilityImpl.ts:66