Introduction

Welcome to RagAPI Docs! API Endpoint: https://api.ragapi.org/ (opens in a new tab)

What is RagAPI?

RagAPI is an API for Retrieval Augmented Generation. Currently we support one endpoint /question where you can ask any document .PDF or .TXT a question.

With RagAPI you can easily build sites like https://pdf.ai/ (opens in a new tab) with only 1 API

Getting Started

API Documentation (opens in a new tab)

import requests
from requests_toolbelt.multipart.encoder import MultipartEncoder
 
def fetchAnswerFromAPI(question, rawFile):
    try:
        url = 'https://api.ragapi.org/question'
        headers = {
            'Authorization': 'Bearer YOUR_RAGAPI_KEY',
        }
 
        multipart_data = MultipartEncoder(
            fields={
                'file': ('file', rawFile),
                'question': question,
                'openai_key': 'YOUR_OPENAI_KEY',
            }
        )
 
        headers['Content-Type'] = multipart_data.content_type
 
        response = requests.post(url, headers=headers, data=multipart_data)
 
        if not response.ok:
            raise Exception(f'API request failed with status {response.status_code}')
 
        json_data = response.json()
 
        if 'answer' in json_data:
            return json_data['answer']
        else:
            return 'No answer found for this question.'
 
    except Exception as error:
        print('API request failed:', error)
        return 'Failed to retrieve an answer. Please try again.'

Chat with PDF