The Vedaya platform provides comprehensive data ingestion capabilities for building your knowledge base.

​
Overview

Data ingestion is the foundation of your RAG system. Vedaya supports multiple file formats and provides various upload methods.

​
Key Features

Multiple Formats

Support for PDF, DOCX, TXT, MD, CSV, JSON and more

Batch Processing

Upload and process multiple files simultaneously

Text Insertion

Direct text input without file upload

Pipeline Monitoring

Real-time processing status tracking

​
Quick Start

​
Text Upload (Recommended)

The simplest way to upload content is via the text endpoint: 
import requests

# Setup
BASE_URL = "https://vedaya-kge.fly.dev"
headers = {"Content-Type": "application/json"}
# Authorization is optional - API works without auth

# Upload text documents
response = requests.post(
    f"{BASE_URL}/documents/texts",
    headers=headers,
    json={
        "texts": [
            "Your document content here...",
            "Another document content..."
        ],
        "file_sources": ["doc1.txt", "doc2.txt"]  # Optional
    }
)

if response.status_code == 200:
    print("✅ Documents uploaded successfully")

​
File Upload (Alternative)

For PDF and other file formats: 
# Upload a file
with open('document.pdf', 'rb') as f:
    files = {'file': f}
    response = requests.post(
        f"{BASE_URL}/documents/upload", 
        headers=headers, 
        files=files
    )

​
Processing Pipeline

1

Upload

Submit files via API
2

Extract

Text extraction from documents
3

Chunk

Split into processable segments
4

Analyze

Entity and relationship extraction
5

Index

Build knowledge graph and vectors

​
Learn More

For detailed documentation on all upload methods and options:

Document Management Guide

Complete guide to document upload, processing, and management

​
API Reference