Skip to main content

​
Installation

npm install @ai-billing/openai-compatible @ai-billing/core @ai-sdk/openai-compatible ai

​
Overview

The @ai-billing/openai-compatible package provides middleware for tracking token usage and calculating costs when using any OpenAI-compatible provider (like xAI, Together AI, or Groq) with the Vercel AI SDK. It captures OpenAI-compatible metrics, such as internalReasoningTokens and inputCacheReadTokens, ensuring that costs are accurately reflected across different providers.

​
Usage

To use the middleware, wrap your OpenAI-compatible model using wrapLanguageModel from the ai package and pass the createOpenAICompatibleMiddleware.
1

Initialize the OpenAI Compatible provider

First, set up the provider using createOpenAICompatible. You must specify a name which will be used as the providerId in the middleware.
import { createOpenAICompatible } from '@ai-sdk/openai-compatible';

const xai = createOpenAICompatible({
  name: 'xai', // must match the providerId used in the middleware
  baseURL: 'https://api.x.ai/v1',
  apiKey: process.env.XAI_API_KEY,
});
2

Define model pricing

Set up a price resolver to define the costs for the models you’ll be using. You can specify costs for standard prompt/completion tokens, as well as provider-specific metrics like internalReasoningTokens or inputCacheReadTokens.
import { createObjectPriceResolver } from '@ai-billing/core';

const priceResolver = createObjectPriceResolver({
  'grok-4-1-fast-reasoning': {
    promptTokens: 0.0000002,
    completionTokens: 0.0000005,
    internalReasoningTokens: 0,
    inputCacheReadTokens: 0.00000015,
  },
});
3

Create the billing middleware

Initialize the OpenAI Compatible billing middleware. You need to provide a destination (such as consoleDestination), your priceResolver, and the providerId that matches the name you used when creating the provider.
import { createOpenAICompatibleMiddleware } from '@ai-billing/openai-compatible';
import { consoleDestination } from '@ai-billing/core';

const billingMiddleware = createOpenAICompatibleMiddleware({
  destinations: [consoleDestination()],
  priceResolver: priceResolver,
  providerId: 'xai', // Must match the provider name used in createOpenAICompatible
});
4

Wrap the model

Use wrapLanguageModel from the ai package to apply the billing middleware to your OpenAI-compatible model.
import { wrapLanguageModel } from 'ai';

const wrappedModel = wrapLanguageModel({
  model: xai('grok-4-1-fast-reasoning'),
  middleware: billingMiddleware,
});
5

Use the wrapped model

Finally, use the wrapped model with AI SDK functions like generateText or streamText. The billing middleware will automatically track tokens, handle provider-specific metrics, and calculate costs.
import { generateText } from 'ai';

const result = await generateText({
  model: wrappedModel,
  prompt: 'What is the capital of Sweden?',
});