VectorStore | DeepFellow Docs

VectorStore

Property	Type	Required	Description
`project_id`	string	Yes	A 24-character hexadecimal MongoDB ObjectId string.
`object`	string	No	The object type, which is always `vector_store`. Always `vector_store`. Default: `"vector_store"`.
`name`	`VectorStoreName`	No	Default: `""`.
`status`	`VectorStoreStatus`	No	Default: `"completed"`.
`file_counts`	`FileCounts`	No	Detailed counts of files by status. Default: `{"cancelled":0,"completed":0,"failed":0,"in_progress":0,"total":0}`.
`expires_after`	`ExpirationPolicy` \| null	No	The expiration policy for a vector store.
`metadata`	`Metadata-Output`	No	Set of key-value pairs that can be attached to an object. Default: `{}`.
`chunking_strategy`	`ChunkingStrategy`	No	The chunking strategy used to chunk the file(s). If not set, will use the auto strategy.Only applicable if file_ids is non-empty. Default: `{"type":"auto","static":{"chunk_overlap_tokens":400,"max_chunk_size_tokens":800}}`.
`knowledge_compression`	`KnowledgeCompressionPolicy` \| null	No	Knowledge compression policy: whether Nightshift stores a summary of each file as vectors in this store.
`knowledge_graph`	`KnowledgeGraphPolicy` \| null	No	Knowledge graph policy: whether Nightshift ingests each processed file into the project's knowledge graph.
`id`	string	Yes	The vector store identifier, which can be referenced in the API endpoints.
`last_active_at`	integer	Yes	The Unix timestamp (in seconds) for when the vector store was last active.
`created_at`	integer	Yes	The Unix timestamp (in seconds) for when the vector store was created.
`expires_at`	integer	Yes	The Unix timestamp (in seconds) for when the vector store will expire.
`bytes`	`UsageBytes`	No	Default: `0`.
`embedding_config`	`EmbeddingConfig`	Yes	Embedding model and size used to create vectors in this vector store.

We use cookies on our website. We use them to ensure proper functioning of the site and, if you agree, for purposes such as analytics, marketing, and targeting ads.