← Blog

2026-10-15 · 5 min read

How Developers Can Access AI Inference Through Air Credits

This article explores how developers can effectively buy Air Credits, enable AI models, and call the /api/v1 endpoint for chat completions in the Air Inference marketplace.

Introduction

The rapid evolution of artificial intelligence has opened new horizons for developers looking to integrate AI capabilities into their applications. Air Inference is a two-sided marketplace that facilitates AI inference by allowing developers to access a variety of OpenAI-compatible APIs. In this article, we will explore how developers can effectively purchase Air Credits, enable AI models, and utilize the /api/v1 endpoint for chat completions in the Air Inference marketplace.

Understanding Air Credits

What are Air Credits?

Air Credits are the currency used within the Air Inference platform. Developers can purchase these credits to access a variety of AI inference services offered by providers on the marketplace. Each service or API call will cost a certain number of Air Credits, which makes it essential for developers to understand how to buy and manage these credits effectively.

Why Use Air Credits?

Using Air Credits simplifies the transaction process for developers. Instead of dealing with complex payment structures for each API call, developers can purchase credits in bulk and use them as needed. This flexibility allows for easier budgeting and management of costs associated with AI inference.

How to Purchase Air Credits

Step 1: Sign Up for an Account

To get started with Air Inference, you'll need to create an account. Visit the Air Inference website and click on the sign-up button. Fill in the required details, such as your email and password, and verify your account through your email.

Step 2: Top Up Your Air Credits

Once you have created an account and logged in, navigate to the "Account" or "Billing" section of the dashboard. Here, you will find an option to buy Air Credits.

  1. Select the Amount: Choose the amount of Air Credits you want to purchase. It's often a good idea to start with a smaller amount if you're testing the platform.

  2. Payment Method: Select your preferred payment method. As of now, Air Inference does not support Stripe Connect payouts, but you can use traditional payment methods available on the platform.

  3. Confirm Purchase: Review your selection and confirm the purchase. Your Air Credits will be credited to your account shortly after the transaction is completed.

Step 3: Monitor Your Air Credits

After purchasing, you can monitor your Air Credits from your account dashboard. This will give you a clear overview of how many credits you have left and how they are being utilized. Keeping track of your credits is essential for managing costs effectively, especially when integrating AI into applications.

Enabling AI Models

Selecting AI Models

Once you have your Air Credits, the next step is to enable the AI models you want to work with. Air Inference hosts a variety of OpenAI-compatible endpoints, including models like vLLM and llama.cpp wrappers.

  1. Browse Available Models: In the marketplace, you can browse through the available AI models. Each model comes with specific capabilities and pricing, so it’s important to review these details before making a selection.

  2. Select Your Model: Once you've identified a model that fits your needs, you can enable it for use. Some models may require additional configuration, so be sure to read the documentation provided by the model’s provider.

Configuring Your Model

After selecting a model, you'll often need to configure it according to your requirements. This may include setting parameters like temperature, max tokens, and prompt format. Make sure to refer to the model's documentation for specific configuration options.

Calling the /api/v1 Endpoint for Chat Completions

Overview of the /api/v1 Endpoint

The /api/v1 endpoint is where developers can send requests for chat completions using the enabled AI models. This endpoint is designed to be compatible with OpenAI's API standards, making it easier for developers familiar with those APIs to transition to Air Inference.

Making Your First API Call

To make a call to the /api/v1 endpoint, you'll need to format your request appropriately. Below is a basic outline of how to structure your API call:

  1. Set the URL: The endpoint URL will be https://api.airinference.com/api/v1.

  2. Select Your Model: Specify which model you want to use in your request.

  3. Prepare Your Request: You will need to send a JSON payload that contains your request parameters, including the prompt and any other settings.

Here’s a brief example of how to structure your API call using curl:

curl -X POST https://api.airinference.com/api/v1 \
-H "Content-Type: application/json" \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{
  "model": "your_model_name",
  "prompt": "What is the significance of AI in modern technology?",
  "max_tokens": 150
}'

Understanding the Response

When you make a successful call to the /api/v1 endpoint, you will receive a JSON response containing the generated chat completion. The response will typically include fields such as the completion text, any warnings, and usage statistics.

It’s important to parse this response correctly in your application to display the generated content effectively to users.

Best Practices for Using Air Inference

Monitor Usage

Keep an eye on how many Air Credits you are using as you make API calls. This will help you stay within budget and optimize your usage.

Experiment with Different Models

Don’t hesitate to try out different models available on the Air Inference marketplace. Each model has unique strengths, and experimenting can help you find the best fit for your specific use case.

Leverage Documentation

Always refer to the official documentation for both Air Inference and the specific models you are using. This will ensure you are utilizing the APIs to their fullest potential and adhering to best practices.

Optimize API Calls

To save on Air Credits, try to optimize your API calls. This could include using shorter prompts or adjusting parameters to get the best results with fewer tokens.

Conclusion

Air Inference offers a streamlined way for developers to access AI inference through the purchase of Air Credits and the use of OpenAI-compatible APIs. By following the steps outlined in this article—purchasing Air Credits, enabling models, and calling the /api/v1 endpoint—developers can efficiently integrate AI capabilities into their applications.

As the AI landscape continues to evolve, platforms like Air Inference will play a crucial role in making these technologies accessible to developers, allowing for innovation and creativity in application development. Whether you are an experienced developer or just starting, utilizing Air Inference can significantly enhance your projects and provide users with powerful AI-driven experiences.