Developer Tools

Build AI Applications With Our API & SDKs

Manage RAG pipelines, process documents, and retrieve information programmatically. Our comprehensive API and native SDKs make it easy to integrate AI capabilities into your applications.

View API Reference Get API Key

# Install the SDK
pip install vectorize

# Initialize client
from vectorize import Client
client = Client(api_key="your_api_key")

# Create a pipeline
pipeline = client.pipelines.create(
  name="product-docs",
  source="google_drive",
  embedding_model="text-embedding-3-small"
)

Developer First

Everything you need to build AI applications

Our API and SDKs provide complete control over your RAG pipelines, from document ingestion to retrieval and evaluation.

RESTful API: Full REST API for managing pipelines, connectors, and retrieving data with automatic query embedding and reranking.
Python SDK: Native Python SDK for seamless integration with your data science and ML workflows.
TypeScript/JavaScript SDK: Modern TypeScript SDK with full type safety for building web applications.
Batch Operations: Efficiently process large volumes of documents with bulk upload and batch retrieval endpoints.
OpenAPI Spec: Complete OpenAPI 3.0 specification for easy client generation in any language.
Advanced Filtering: Query your data with metadata filters, date ranges, and custom field searches.

Quick Start

Get started in minutes

Choose your preferred language and start building with Vectorize.

Python

from vectorize import VectorizeClient

# Initialize the client
client = VectorizeClient(api_key="your-api-key")

# Create a RAG pipeline
pipeline = client.pipelines.create(
    name="product-docs",
    vector_database="pinecone",
    embedding_model="text-embedding-3-small"
)

# Upload documents
client.documents.upload(
    pipeline_id=pipeline.id,
    files=["product-manual.pdf", "user-guide.docx"]
)

# Query with automatic embedding
results = client.retrieve(
    pipeline_id=pipeline.id,
    query="How do I reset my password?",
    num_results=5
)

TypeScript

import { VectorizeClient } from '@vectorize/sdk';

// Initialize the client
const client = new VectorizeClient({ 
  apiKey: process.env.VECTORIZE_API_KEY 
});

// Create a connector
const connector = await client.connectors.create({
  type: 'sharepoint',
  name: 'Corporate Knowledge Base',
  config: {
    tenant_id: 'your-tenant-id',
    client_id: 'your-client-id'
  }
});

// Advanced query with filters
const results = await client.retrieve({
  pipelineId: 'pipeline-123',
  question: 'quarterly sales report',
  advancedQuery: {
    mode: 'hybrid',
    filters: {
      range: { date: { gte: '2024-01-01' } }
    }
  }
});

cURL

# Retrieve documents with automatic query embedding
curl -X POST https://api.vectorize.io/v1/pipelines/{pipeline_id}/retrieve \
  -H "Authorization: Bearer your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "question": "What are the compliance requirements?",
    "numResults": 10,
    "metadata-filters": {
      "document_type": "compliance",
      "year": 2024
    }
  }'

# Create a new RAG pipeline
curl -X POST https://api.vectorize.io/v1/pipelines \
  -H "Authorization: Bearer your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Legal Documents",
    "embedding_model": "text-embedding-3-small",
    "vector_database": "built-in"
  }'

Build anything with our API

From simple search to complex AI workflows, our API powers it all.

Build AI-powered search into your application
Automate document processing workflows
Create custom RAG evaluation pipelines
Integrate with CI/CD for continuous deployment
Develop chatbots with up-to-date knowledge
Power internal tools with semantic search

Start building with Vectorize API

Get your API key and start building AI-powered applications today. Free tier available for developers.

Get started free Read the docs