Developer Guide

Making Your First Request

Now that you understand endpoints and authentication, let's make actual API calls. NeevCloud provides ready-to-use examples in three popular languages.

cURL Example

cURL is a command-line tool available on most systems. It's excellent for testing API endpoints quickly.

curl https://inference.ai.neevcloud.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-nc-xxxxx" \
  -d '{
    "model": "gpt-4o-mini",
    "messages": [
      { "role": "system", "content": "You are a helpful assistant." },
      { "role": "user", "content": "Explain inference API design." }
    ],
    "max_tokens": 100,
    "temperature": 0.7
  }'

Breaking Down the Request

• curl https://inference.ai.neevcloud.com/v1/chat/completions – The HTTP client and endpoint URL

• -H "Authorization: Bearer $NEEV_API_KEY" – Your authentication. Replace $NEEV_API_KEY with your actual key, or set it as an environment variable

• -H "Content-Type: application/json" – Tells the server you're sending JSON data

• -d '{...}' – The request body containing your parameters

Request Parameters

• messages – The conversation history (system, user, assistant roles)

• model – The specific model name you're using. Must match exactly the model you selected

• max_tokens – Maximum number of tokens in the response. Limits the output length and costs

• temperature – Controls randomness (0 = deterministic, 1 = creative)

Response

You'll receive a JSON response containing:

• The generated text

• Token counts (input and output)

• Processing time

• Any metadata about the request

Python Example

Python is ideal for integrating AI into web applications, data pipelines, and automation scripts.

from openai import OpenAI

client = OpenAI(
    api_key="sk-nc-xxxxx",
    base_url="https://inference.ai.neevcloud.com/v1"
)

response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Explain inference API design."}
    ],
    max_tokens=100,
    temperature=0.7
)

print(response.choices[0].message.content)

How This Works

• Import OpenAI – The official Python library for OpenAI-compatible APIs

• Initialize client – Configure the OpenAI client with your API key and NeevCloud base URL

• Make the request – client.chat.completions.create() sends the request and returns a structured object

• Use the data – Access content via response.choices[0].message.content

Error Handling

In production code, add error handling:

from openai import APIError

try:
    response = client.chat.completions.create(
        model="gpt-4o-mini",
        messages=[{"role": "user", "content": "Hello"}],
        max_tokens=100,
        temperature=0.7
    )
    print(response.choices[0].message.content)
except APIError as e:
    print(f"OpenAI API returned an API Error: {e}")
except Exception as e:
    print(f"An unknown error occurred: {e}")

JavaScript Example

JavaScript is essential for Node.js backends and browser-based applications (though API keys should never be exposed in browser code—use a backend proxy instead).

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "sk-nc-xxxxx",
  baseURL: "https://inference.ai.neevcloud.com/v1",
});

const response = await client.chat.completions.create({
  model: "gpt-4o-mini",
  messages: [
    { role: "system", content: "You are a helpful assistant." },
    { role: "user", content: "Explain inference API design." }
  ],
  max_tokens: 100,
  temperature: 0.7,
});

console.log(response.choices[0].message.content);

Understanding the Code

• Import OpenAI – The official Node.js library for OpenAI-compatible APIs

• Initialize client – Configure with API key and base URL

• Make the request – Call client.chat.completions.create() to send data

• Access results – Read the generated text from response.choices[0].message.content

Using in Node.js (Raw HTTP)

If you prefer using fetch instead of the SDK:

async function generateText(prompt) {
  const response = await fetch('https://inference.ai.neevcloud.com/v1/chat/completions', {
    method: 'POST',
    headers: {
      'Authorization': `Bearer ${process.env.NEEV_API_KEY}`,
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({
      messages: [
          { role: "user", content: prompt }
      ],
      model: "Llama 3.1 405B",
      max_tokens: 100,
      temperature: 0.7
    })
  });
  
  const result = await response.json();
  return result;
}

// Use it
const result = await generateText("Explain quantum computing in simple terms");
console.log(result.choices[0].message.content);

Testing Your Setup

After running any of these examples:

• Verify the response – You should receive generated text back

• Check token counts – The response includes input and output token usage

• Review timing – Note how long the request took

• Test error cases – Try an invalid API key to confirm error handling works

Once you've successfully made a request, you're ready to integrate the API into your application.