> ## Documentation Index
> Fetch the complete documentation index at: https://narev.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Getting started

> First steps with Together AI middleware for tracking AI costs.

## Installation

<CodeGroup>
  ```bash npm theme={null}
  npm install @ai-billing/togetherai @ai-billing/core @ai-sdk/togetherai ai
  ```

  ```bash pnpm theme={null}
  pnpm add @ai-billing/togetherai @ai-billing/core @ai-sdk/togetherai ai
  ```

  ```bash yarn theme={null}
  yarn add @ai-billing/togetherai @ai-billing/core @ai-sdk/togetherai ai
  ```
</CodeGroup>

## Overview

The `@ai-billing/togetherai` package provides middleware for tracking token usage and calculating costs when using Together AI models with the Vercel AI SDK.

Together AI's usage payload is a flat, OpenAI-compatible block (`prompt_tokens`, `completion_tokens`, `total_tokens`, `reasoning_tokens`) with no cache fields, so cache-read and cache-write tokens are always billed as zero.

## Usage

To use the middleware, wrap your Together AI model using `wrapLanguageModel` from the `ai` package and pass the `createTogetheraiMiddleware`.

<Steps>
  <Step title="Initialize the Together AI provider">
    First, set up the Together AI provider using your API key.

    ```typescript theme={null}
    import { createTogetherAI } from '@ai-sdk/togetherai';

    const togetherai = createTogetherAI({
      apiKey: process.env.TOGETHER_API_KEY,
    });
    ```
  </Step>

  <Step title="Define model pricing">
    Set up a price resolver to define the costs for the models you'll be using.

    ```typescript theme={null}
    import { createObjectPriceResolver } from '@ai-billing/core';

    const priceResolver = createObjectPriceResolver({
      'openai/gpt-oss-20b': {
        promptTokens: 0.05 / 1_000_000,
        completionTokens: 0.2 / 1_000_000,
      },
    });
    ```
  </Step>

  <Step title="Create the billing middleware">
    Initialize the Together AI billing middleware. You need to provide a destination (such as `consoleDestination`) where billing events will be sent, along with your `priceResolver`.

    ```typescript theme={null}
    import { createTogetheraiMiddleware } from '@ai-billing/togetherai';
    import { consoleDestination } from '@ai-billing/core';

    const billingMiddleware = createTogetheraiMiddleware({
      destinations: [consoleDestination()],
      priceResolver: priceResolver,
    });
    ```
  </Step>

  <Step title="Wrap the model">
    Use `wrapLanguageModel` from the `ai` package to apply the billing middleware to your Together AI model.

    ```typescript theme={null}
    import { wrapLanguageModel } from 'ai';

    const wrappedModel = wrapLanguageModel({
      model: togetherai('openai/gpt-oss-20b'),
      middleware: billingMiddleware,
    });
    ```
  </Step>

  <Step title="Use the wrapped model">
    Finally, use the wrapped model with AI SDK functions like `generateText` or `streamText`. The billing middleware will automatically track tokens and calculate costs.

    ```typescript theme={null}
    import { generateText } from 'ai';

    const result = await generateText({
      model: wrappedModel,
      prompt: 'What is the capital of Sweden?',
    });
    ```
  </Step>
</Steps>