> ## 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.

# CreateGroqV3Middleware

[**@ai-billing/groq**](../index)

***

[@ai-billing/groq](/docs/sdk/ai-billing/reference/groq/typedoc/index) / createGroqV3Middleware

# Function: createGroqV3Middleware()

> **createGroqV3Middleware**\<`TTags`>(`options`): `LanguageModelV3Middleware`

Defined in: [groq/src/ai-sdk/language-model-middleware/v3/language-model-v3-groq-billing-middleware.ts:94](https://github.com/narevai/ai-billing/blob/main/packages/groq/src/ai-sdk/language-model-middleware/v3/language-model-v3-groq-billing-middleware.ts#L94)

Creates a V3 billing middleware for the Groq provider (`@ai-sdk/groq`).
Derives token usage from Groq's raw usage payload and resolves cost from pricing plus usage.

## Type Parameters

### TTags

`TTags` *extends* `JSONObject`

## Parameters

### options

[`GroqV3MiddlewareOptions`](/docs/sdk/ai-billing/reference/groq/typedoc/interfaces/GroqV3MiddlewareOptions)\<`TTags`>

Billing options; see [GroqV3MiddlewareOptions](/docs/sdk/ai-billing/reference/groq/typedoc/interfaces/GroqV3MiddlewareOptions). A `priceResolver` is required
because Groq does not supply billed amounts in provider metadata the way the AI Gateway does.

## Returns

`LanguageModelV3Middleware`

A V3 billing middleware instance for Groq.

## Example

Same wiring as `examples/dev-sandbox/app/api/groq` (`createGroqMiddleware` is this function's export alias
from `@ai-billing/groq`).

```ts theme={null}
import { createGroq } from '@ai-sdk/groq';
import { wrapLanguageModel } from 'ai';
import { createGroqMiddleware } from '@ai-billing/groq';
import {
  consoleDestination,
  createObjectPriceResolver,
  type ModelPricing,
} from '@ai-billing/core';

const groq = createGroq({ apiKey: process.env.GROQ_API_KEY });

const customPricingMap: Record<string, ModelPricing> = {
  'openai/gpt-oss-120b': {
    promptTokens: 0.15 / 1_000_000,
    completionTokens: 0.6 / 1_000_000,
    inputCacheReadTokens: 0.075 / 1_000_000,
    inputCacheWriteTokens: 0,
  },
};

const priceResolver = createObjectPriceResolver(customPricingMap);

const billingMiddleware = createGroqMiddleware({
  destinations: [consoleDestination()],
  priceResolver,
});

const wrappedModel = wrapLanguageModel({
  model: groq('openai/gpt-oss-120b'),
  middleware: billingMiddleware,
});
```