← Blog

2026-11-15 · 5 min read

How Developers Access AI Inference Through Air Credits

Explore how developers can top up Air Credits, enable AI models, and effectively call the /api/v1 chat completions endpoint in the Air Inference marketplace.

Introduction to Air Inference

The rise of artificial intelligence (AI) has opened a plethora of opportunities for developers and businesses looking to harness its power. Air Inference is a two-sided marketplace that bridges the gap between developers seeking AI inference and providers offering AI models and endpoints. This platform allows developers to top up Air Credits, enabling them to seamlessly call OpenAI-compatible APIs. In this article, we will explore how developers can effectively access AI inference through Air Credits, enabling AI models, and calling the /api/v1 chat completions endpoint.

Understanding Air Credits

Air Credits are the currency used within the Air Inference marketplace. They empower developers to access various AI models offered by providers. By topping up their Air Credits, developers can easily make API calls to the models listed on the platform.

How to Buy Air Credits

Purchasing Air Credits is a straightforward process. Here are the steps developers can follow:

  1. Create an Account: First, you need to register on the Air Inference platform. Visit the Air Inference website and sign up for an account.

  2. Navigate to the Top-Up Section: Once your account is created, log in and navigate to the section where you can top up your Air Credits. This is typically located in your account settings or dashboard.

  3. Select the Amount: Choose how many Air Credits you want to purchase. The platform generally offers various tiers to accommodate different usage needs.

  4. Payment Method: Enter your payment details. Currently, transactions are processed off-platform, so make sure to follow the instructions provided for a smooth transaction.

  5. Confirm the Purchase: After entering the necessary details, confirm your purchase. Once the transaction is successful, your Air Credits will be reflected in your account balance.

Managing Your Air Credits

After topping up, it’s essential to manage your Air Credits wisely. Keep track of your usage, as each API call will deduct credits based on the model's pricing. The dashboard provides insights into your current balance and usage history, allowing you to make informed decisions regarding your future purchases.

Enabling AI Models

Once you have topped up your Air Credits, the next step is to enable AI models that you wish to use. Air Inference offers a variety of models from different providers, such as vLLM or llama.cpp wrappers, which are compatible with OpenAI APIs.

How to Enable Models

  1. Browse Available Models: In your developer dashboard, explore the list of available AI models. Providers list their endpoints along with documentation regarding their capabilities, inputs, and expected outputs.

  2. Select a Model: Once you find a model that fits your project requirements, select it. This action usually takes you to a page with more details about the model, including its pricing and usage instructions.

  3. Check Compatibility: Ensure that the model is OpenAI-compatible. This step is crucial, as Air Inference primarily focuses on APIs that conform to OpenAI’s standards, allowing for easier integration.

  4. Enable the Model: Follow the prompts to enable the selected model. This may involve agreeing to terms of service or confirming your understanding of the model’s capabilities.

  5. Access Endpoint: Once the model is enabled, you will receive the endpoint details necessary for making API calls. The common endpoint for chat completions is /api/v1.

Calling the /api/v1 Chat Completions Endpoint

With your Air Credits topped up and models enabled, you are now ready to call the /api/v1 chat completions endpoint. This API endpoint allows developers to generate conversational responses based on user inputs.

Making API Calls

To make a successful API call to the /api/v1 endpoint, you will need to include certain parameters, such as the input prompt and any specific configurations for the model you are using.

Here’s a general outline of the steps to call the /api/v1 chat completions endpoint:

  1. Prepare Your Request: You will need to structure your API request according to the specifications provided by the model you selected. This includes setting up headers, specifying the model name, and including the input prompt.

  2. Set Up Authentication: Most API calls require some form of authentication. Make sure you have the necessary API key or token to authenticate your requests.

  3. Send the Request: Use your preferred method (like cURL, Postman, or a programming language library) to send the request to the API endpoint.

  4. Handle the Response: Once you receive a response from the API, process it according to your application's requirements. This may involve parsing JSON, displaying the output on a user interface, or further manipulating the data.

Example of an API Call

Here’s a short example of how you might structure a cURL command to call the /api/v1 chat completions endpoint:

curl -X POST https://api.airinference.com/api/v1/chat/completions \
     -H "Content-Type: application/json" \
     -H "Authorization: Bearer YOUR_API_KEY" \
     -d '{
           "model": "MODEL_NAME",
           "messages": [{"role": "user", "content": "Hello, how can I use Air Inference?"}]
         }'

In this example, replace YOUR_API_KEY with your actual API key and MODEL_NAME with the name of the model you are using. The request includes a message from the user, which the model will respond to.

Best Practices for Using Air Inference

To maximize your productivity and efficiency while using Air Inference, consider following these best practices:

  1. Understand Model Limitations: Each model has its strengths and weaknesses. Familiarize yourself with the capabilities of the models you choose to avoid using them for tasks they are not suited for.

  2. Monitor Usage: Regularly check your Air Credits balance and usage statistics. This practice helps you stay within budget and plan for future top-ups.

  3. Review Documentation: Always refer to the model’s documentation for specific details on parameters, limitations, and best practices for optimal usage.

  4. Experiment with Prompts: When using chat completion models, the quality of your input prompts can significantly affect the output. Experiment with different prompts to achieve the best results.

  5. Engage with the Community: Join forums and communities related to Air Inference and AI development. Sharing experiences, tips, and resources can enhance your understanding and usage of the platform.

Conclusion

Air Inference offers a powerful and flexible solution for developers looking to access AI inference through a marketplace model. By topping up Air Credits, enabling models, and effectively calling the /api/v1 chat completions endpoint, developers can seamlessly integrate AI capabilities into their applications.

As the landscape of AI continues to evolve, platforms like Air Inference will play an essential role in making AI more accessible to developers. By following the steps outlined in this guide, you can leverage the full potential of AI inference in your projects. Whether you're developing chatbots, content generators, or other AI-driven applications, Air Inference provides the tools and resources you need to succeed.