Class LibSQLVectorStore<Metadata>

constructor

• new LibSQLVectorStore<Metadata extends MetadataDefault = MetadataDefault>(
      embeddings: EmbeddingsInterface,
      options: LibSQLVectorStoreArgs,
  ): LibSQLVectorStore<Metadata>

  Initializes a new instance of the LibSQLVectorStore.

  Type Parameters

  • Metadata extends MetadataDefault = MetadataDefault

  Parameters

  • embeddings: EmbeddingsInterface

    The embeddings interface to use.

  • options: LibSQLVectorStoreArgs

    Configuration options for the vector store.

  Returns LibSQLVectorStore<Metadata>

FilterType

_vectorstoreType

• _vectorstoreType(): string

  Returns the type of vector store.

  Returns string

  The string "libsql".

addDocuments

• addDocuments(documents: Document<Metadata>[]): Promise<string[]>

  Adds documents to the vector store.

  Parameters

  • documents: Document<Metadata>[]

    The documents to add.

  Returns Promise<string[]>

  The IDs of the added documents.

addVectors

• addVectors(
      vectors: number[][],
      documents: Document<Metadata>[],
  ): Promise<string[]>

  Adds vectors to the vector store.

  Parameters

  • vectors: number[][]

    The vectors to add.

  • documents: Document<Metadata>[]

    The documents associated with the vectors.

  Returns Promise<string[]>

  The IDs of the added vectors.

delete

• delete(
      params: { deleteAll?: boolean; ids?: string[] | number[] },
  ): Promise<void>

  Deletes vectors from the store.

  Parameters

  • params: { deleteAll?: boolean; ids?: string[] | number[] }

    Delete parameters.

    • OptionaldeleteAll?: boolean

    • Optionalids?: string[] | number[]

      The ids of the vectors to delete.

  Returns Promise<void>

similaritySearchVectorWithScore

• similaritySearchVectorWithScore(
      query: number[],
      k: number,
      filter?: InStatement | WhereCondition<Metadata>,
  ): Promise<[Document<Metadata>, number][]>

  Performs a similarity search using a vector query and returns documents with their scores.

  Parameters

  • query: number[]

    The query vector.

  • k: number

    The number of results to return.

  • Optionalfilter: InStatement | WhereCondition<Metadata>

  Returns Promise<[Document<Metadata>, number][]>

  An array of tuples containing the similar documents and their scores.

StaticfromDocuments

• fromDocuments<Metadata extends MetadataDefault = MetadataDefault>(
      docs: Document<Metadata>[],
      embeddings: EmbeddingsInterface,
      options?: LibSQLVectorStoreArgs,
  ): Promise<LibSQLVectorStore<Metadata>>

  Creates a new LibSQLVectorStore instance from documents.

  Type Parameters

  • Metadata extends MetadataDefault = MetadataDefault

  Parameters

  • docs: Document<Metadata>[]

    The documents to add to the store.

  • embeddings: EmbeddingsInterface

    The embeddings interface to use.

  • Optionaloptions: LibSQLVectorStoreArgs

    Configuration options for the vector store.

  Returns Promise<LibSQLVectorStore<Metadata>>

  A new LibSQLVectorStore instance.

StaticfromTexts

• fromTexts<Metadata extends MetadataDefault = MetadataDefault>(
      texts: string[],
      metadatas: Metadata | Metadata[],
      embeddings: EmbeddingsInterface,
      options?: LibSQLVectorStoreArgs,
  ): Promise<LibSQLVectorStore<Metadata>>

  Creates a new LibSQLVectorStore instance from texts.

  Type Parameters

  • Metadata extends MetadataDefault = MetadataDefault

  Parameters

  • texts: string[]

    The texts to add to the store.

  • metadatas: Metadata | Metadata[]

    The metadata for the texts.

  • embeddings: EmbeddingsInterface

    The embeddings interface to use.

  • Optionaloptions: LibSQLVectorStoreArgs

    Configuration options for the vector store.

  Returns Promise<LibSQLVectorStore<Metadata>>

  A new LibSQLVectorStore instance.