tasks-retrieval package

Classes

Class	Description
DefaultTextChunker	WASM-backed text chunker supporting sentence-aware character and word chunking.
FilesetResolver	Resolves the files required for the MediaPipe Task APIs.This class verifies whether SIMD is supported in the current environment and loads the SIMD files only if support is detected. The returned filesets require that the Wasm files are published without renaming. If this is not possible, you can invoke the MediaPipe Tasks APIs using a manually created WasmFileset.
MemoryVectorStore	In-memory vector store implementing cosine-similarity nearest-neighbor search and metadata filtering.
SemanticRetriever	Multimodal semantic retrieval task for indexing and querying text, image, and audio documents.
SemanticRetrieverComponents	Builder and container for configuring SemanticRetriever embedding providers, vector store, and text chunker.
UniversalEmbedder	Multimodal universal embedder supporting text, image, audio, and interleaved content embeddings.

Interfaces

Interface	Description
Embedding	List of embeddings with an optional timestamp.One and only one of the two 'floatEmbedding' and 'quantizedEmbedding' will contain data, based on whether or not the embedder was configured to perform scalar quantization.
RetrievalRecord	A document or chunk record stored in a vector store, containing multimodal content, embeddings, and metadata.
RetrievalResult	A single semantic search result returned by SemanticRetriever.retrieve().
SemanticRetrieverOptions	Options for configuring SemanticRetriever and retrieval queries (also exported as RetrievalOptions).
UniversalEmbedderOptions	Options to configure the MediaPipe Universal Embedder Task.
UniversalEmbedderResult	Embedding results returned by UniversalEmbedder.
WasmFileset	A configuration object that provides the location of the Wasm binary and its loader.

Type Aliases

Type Alias	Description
ChunkingMode	Text chunking granularity mode ('CHARACTER' or 'WORD').
TaskPart	A single multimodal content part (text, image, or audio; also exported as ContentPart).
TaskPartKind	Discriminator kind for a TaskPart ('TEXT', 'IMAGE', or 'AUDIO').