Getting Started

Obtaining an API key

Every request to PurAI is authenticated with an API key. To create one:

Create a PurAI account and open your account dashboard
Navigate to the API Keys section
Select Create key
Your key is displayed once. Store it securely, and never share it or commit it to source control.

Making your first request

You interact with PurAI through standard HTTP POST requests — from cURL, fetch(), or any HTTP client — which makes it easy to use from any platform or language.

A simple demonstration using the chat/completions endpoint is below:

More info on our packages, PurAI.js and PurAI.py, can be found in our packages page!

PurAI.js (NodeJS)

index.js


const PurAI = require('PurAI.js');
const { OpenAIChatCompletionsMessageRole, OpenAIChatCompletionsModel } = require('PurAI.js/src/types');
 
const PurAI = new PurAI('YOUR API KEY');
 
let response = await PurAI.openAI.chat.completions({
    messages: [
        {
            role: OpenAIChatCompletionsMessageRole.User,
            content: 'Hello, PurAI!'
        }
    ],
    model: OpenAIChatCompletionsModel['GPT-3.5 Turbo'] // ANY DESIRED MODEL
});
 
console.log(response);

PurAI.js (using import)

index.ts


import PurAI from 'PurAI.js';
import { OpenAIChatCompletionsMessageRole, OpenAIChatCompletionsModel } from 'PurAI.js/src/types';
 
const PurAI = new PurAI('YOUR API KEY');
 
let response = await PurAI.openAI.chat.completions({
    messages: [
        {
            role: OpenAIChatCompletionsMessageRole.User,
            content: 'Hello, PurAI!'
        }
    ],
    model: OpenAIChatCompletionsModel['GPT-3.5 Turbo'] // ANY DESIRED MODEL
});
 
console.log(response);

PurAI.py

main.py


import PurAI
import os
 
token = "YOUR API KEY" 
 
ai = PurAI.PurAI(token)
 
print(ai.generate("Say hello"))

JavaScript/Ajax

index.js


const url = 'https://ai.purlabs.xyz/openai/chat/completions';
 
const data = JSON.stringify({
    "messages": [{ "role": "user", "content": "Hello, PurAI!" }],
    "model": "gpt-3.5-turbo" // ANY DESIRED MODEL
});
 
const response = await fetch(url, {
    method: 'POST',
    headers: {
        'Content-Type': 'application/json',
        'Authorization': 'Bearer YOUR API KEY',
    },
    body: data,
});
 
const text = await response.text();
 
console.log(text);

Node.js

index.js


const http = require("https");
 
const url = 'https://ai.purlabs.xyz/v1/chat/completions';
 
const options = {
    method: 'POST',
    'Content-Type': 'application/json',
    'Authorization': 'Bearer YOUR API KEY',
};
 
const data = JSON.stringify({
    "messages": [{ "role": "user", "content": "Hello, PurAI!" }],
    "model": "gpt-3.5-turbo" // ANY DESIRED MODEL
});
 
let result = '';
const req = http.request(url, options, (res) => {
    console.log(res.statusCode);
 
    res.setEncoding('utf8');
    res.on('data', (chunk) => {
        result += chunk;
    });
 
    res.on('end', () => {
        console.log(result);
    });
});
 
req.on('error', (e) => {
    console.error(e);
});
 
req.write(data);
req.end();

Python

index.py


import requests
from requests.structures import CaseInsensitiveDict
 
url = "https://ai.purlabs.xyz/v1/chat/completions"
 
headers = CaseInsensitiveDict()
headers["Content-Type"] = "application/json"
headers["Authorization"] = "Bearer YOUR API KEY"
 
data = """
{
    "messages": [{ "role": "user", "content": "Hello, PurAI!" }],
    "model": "gpt-3.5-turbo" // ANY DESIRED MODEL
}
"""
 
resp = requests.post(url, headers=headers, data=data)
 
print(resp.status_code)

C#

index.cs


var url = "https://ai.purlabs.xyz/v1/chat/completions";
 
var httpRequest = (HttpWebRequest)WebRequest.Create(url);
 
httpRequest.Method = "POST";
httpRequest.Headers["Authorization"] = "Bearer YOUR API KEY";
httpRequest.ContentType = "application/json";
 
var data = @"{
        ""messages"": [{ ""role"": ""user"", ""content"": ""Hello, PurAI!"" }],
        ""model"": ""gpt-3.5-turbo"" // ANY DESIRED MODEL
    }";

cURL/BASH

Console


curl -X POST https://ai.purlabs.xyz/v1/chat/completions -H "Content-Type: application/json" -H "Authorization: Bearer YOUR API KEY" --data-binary @- <<DATA
    {
        "messages": [{ "role": "user", "content": "Hello, PurAI!" }],
        "model": "gpt-3.5-turbo" // ANY DESIRED MODEL
    }
DATA

Returns:

Console


{
 "id": "chatcmpl-7aW8BDCAXErPoazm2dSI1EKEUzKBQ",
 "object": "chat.completion",
 "created": 1688937675,
 "model": "gpt-3.5-turbo-0613",
 "choices": [{
     "index": 0,
     "message": {
         "role": "assistant",
         "content": "Hello! How can I assist you today?"
     },
     "finish_reason": "stop"
 }],
 "usage": {
     "prompt_tokens": 13,
     "completion_tokens": 9,
     "total_tokens": 22
 }
}

How PurAI is structured

PurAI is simple: you send a POST request to a URL that combines your desired company (openai, for example) and product (chat/completions, for example), and receive the model’s output in response.

The full list of URLs, along with their descriptions, is documented in the following sections.