> ## Documentation Index
> Fetch the complete documentation index at: https://docs.runflow.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Create vector store (CLI/SDK)

> Creates a new vector store with specified embedding configuration.



## OpenAPI

````yaml /api-reference/openapi.json post /api/v1/runtime/vector-stores
openapi: 3.0.0
info:
  title: Runflow Runtime API
  description: >-
    Programmatic surface for the Runflow platform — used by the SDK, the CLI,
    the execution engine, and any server-to-server caller. Authenticate with an
    `x-api-key` header (or `Authorization: Bearer ...`).
  version: 1.0.0
  contact: {}
servers:
  - url: https://api.runflow.ai
    description: Production
  - url: http://localhost:3000
    description: Local development
security: []
tags: []
paths:
  /api/v1/runtime/vector-stores:
    post:
      tags:
        - Runtime API - Vector Stores Management
      summary: Create vector store (CLI/SDK)
      description: Creates a new vector store with specified embedding configuration.
      operationId: RuntimeVectorStoresController_createVectorStore
      parameters: []
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/CreateVectorStoreDto'
      responses:
        '201':
          description: Vector store created successfully.
      security:
        - bearer: []
components:
  schemas:
    CreateVectorStoreDto:
      type: object
      properties:
        name:
          type: string
          description: Vector store name
        description:
          type: string
          description: Description
        embeddingConfigId:
          type: string
          description: >-
            Embedding configuration ID (optional if llmProviderConfigId +
            llmModelConfigId provided)
        llmProviderConfigId:
          type: string
          description: LLM Provider Config ID (for tenant-credential embeddings)
        llmModelConfigId:
          type: string
          description: LLM Model Config ID (for tenant-credential embeddings)
        type:
          type: string
          description: Vector store type
          enum:
            - KNOWLEDGE
            - STRUCTURED
          default: KNOWLEDGE
        metadata:
          type: object
          description: Custom metadata
        structuredType:
          type: string
          description: Structured document type (only for STRUCTURED stores)
          enum:
            - FAQ
            - SURVEY
            - GLOSSARY
            - CATALOG
            - QA
            - DEFINITION
        chunkSize:
          type: number
          description: 'Chunk size in tokens (default: 1000)'
          minimum: 100
          maximum: 8000
          default: 1000
        chunkOverlap:
          type: number
          description: 'Chunk overlap in tokens (default: 200)'
          minimum: 0
          maximum: 1000
          default: 200
        chunkStrategy:
          type: string
          description: 'Chunking strategy (default: recursive)'
          enum:
            - fixed
            - recursive
            - semantic
          default: recursive
      required:
        - name

````