Implementation:Openai Openai node VectorStore Files

Knowledge Sources	Openai_Openai_node
Domains	SDK, Vector_Stores, File_Management
Last Updated	2026-02-15 12:00 GMT

Overview

The Files resource class provides methods for managing individual files within an OpenAI vector store, including creating, retrieving, updating, listing, deleting, uploading, polling, and reading file content.

Description

The Files class extends APIResource and represents the individual file management layer within a vector store. Each file attached to a vector store is asynchronously processed for embedding and indexing, and this class provides both the low-level CRUD endpoints and higher-level convenience methods for common workflows.

The core API methods (create, retrieve, update, list, delete, content) map directly to the OpenAI REST API endpoints under /vector_stores/{id}/files. The update method allows modifying file attributes after attachment, and the content method retrieves the parsed text content of a processed file.

The convenience methods (createAndPoll, poll, upload, uploadAndPoll) simplify asynchronous workflows. The poll method loops until a file reaches a terminal status (completed or failed), respecting a configurable or server-suggested polling interval. The upload method first uploads a file to the Files API with purpose assistants and then attaches it to the vector store. All endpoints include the OpenAI-Beta: assistants=v2 header.

Usage

Use this resource when you need to manage individual files within a vector store -- attaching new files, checking their processing status, reading their parsed content, or removing them. For bulk file operations, use the related FileBatches resource instead.

Code Reference

Source Location

• Repository: openai-node
• File: src/resources/vector-stores/files.ts

Signature

class Files extends APIResource {
  create(vectorStoreID: string, body: FileCreateParams, options?: RequestOptions): APIPromise<VectorStoreFile>;
  retrieve(fileID: string, params: FileRetrieveParams, options?: RequestOptions): APIPromise<VectorStoreFile>;
  update(fileID: string, params: FileUpdateParams, options?: RequestOptions): APIPromise<VectorStoreFile>;
  list(vectorStoreID: string, query?: FileListParams, options?: RequestOptions): PagePromise<VectorStoreFilesPage, VectorStoreFile>;
  delete(fileID: string, params: FileDeleteParams, options?: RequestOptions): APIPromise<VectorStoreFileDeleted>;
  createAndPoll(vectorStoreId: string, body: FileCreateParams, options?: RequestOptions & { pollIntervalMs?: number }): Promise<VectorStoreFile>;
  poll(vectorStoreID: string, fileID: string, options?: RequestOptions & { pollIntervalMs?: number }): Promise<VectorStoreFile>;
  upload(vectorStoreId: string, file: Uploadable, options?: RequestOptions): Promise<VectorStoreFile>;
  uploadAndPoll(vectorStoreId: string, file: Uploadable, options?: RequestOptions & { pollIntervalMs?: number }): Promise<VectorStoreFile>;
  content(fileID: string, params: FileContentParams, options?: RequestOptions): PagePromise<FileContentResponsesPage, FileContentResponse>;
}

Import

import OpenAI from 'openai';

I/O Contract

Inputs

Name	Type	Required	Description
vectorStoreID	string	Yes	The ID of the vector store.
body.file_id	string	Yes	The File ID to attach to the vector store.
body.attributes	number | boolean> | null	No	Up to 16 key-value metadata pairs.
body.chunking_strategy	FileChunkingStrategyParam	No	Chunking strategy: auto or static.
params.vector_store_id	string	Yes	Vector store ID (used for retrieve, update, delete, content).
query.filter	'completed' | 'failed' | 'cancelled'	No	Filter files by processing status.
query.order	'desc'	No	Sort order by created_at.
options.pollIntervalMs	number	No	Custom poll interval in ms (default: 5000 or server-suggested).

Outputs

Name	Type	Description
id	string	The vector store file identifier.
created_at	number	Unix timestamp (seconds) when the file was created.
status	'completed' | 'cancelled' | 'failed'	Processing status.
usage_bytes	number	Total bytes used by the file in the vector store.
vector_store_id	string	The parent vector store ID.
last_error	null	Error details if processing failed.
attributes	number | boolean> | null	Metadata key-value pairs.
chunking_strategy	FileChunkingStrategy	The chunking strategy used for the file.

Usage Examples

import OpenAI from 'openai';
import fs from 'fs';

const client = new OpenAI();

// Upload a file and poll until it is fully processed
const file = await client.vectorStores.files.uploadAndPoll(
  'vs_abc123',
  fs.createReadStream('knowledge-base.pdf'),
);
console.log(file.status); // 'completed'

// Retrieve file details
const fileInfo = await client.vectorStores.files.retrieve(
  'vsfile_abc123',
  { vector_store_id: 'vs_abc123' },
);

// Update file attributes
await client.vectorStores.files.update(
  'vsfile_abc123',
  { vector_store_id: 'vs_abc123', attributes: { category: 'legal', priority: 1 } },
);

// List all completed files in a vector store
const completedFiles = await client.vectorStores.files.list(
  'vs_abc123',
  { filter: 'completed', order: 'desc' },
);

// Read parsed content of a processed file
const content = await client.vectorStores.files.content(
  'vsfile_abc123',
  { vector_store_id: 'vs_abc123' },
);

// Delete a file from the vector store (does not delete the underlying file)
await client.vectorStores.files.delete(
  'vsfile_abc123',
  { vector_store_id: 'vs_abc123' },
);

Related Pages

• Environment:Openai_Openai_node_Node_20_Runtime