Custom authentication
Use custom credentials instead of relying on the default environment variable.
AI Tools · Function
createGroq
Creates a customized Groq provider for the Vercel AI SDK. Configure credentials and the API endpoint, then call the provider with a Groq model identifier.
How it works
createGroq builds a Groq provider instance from optional settings. The resulting provider can be called with a model identifier to obtain a language model, while configuration can customize credentials and the API endpoint prefix.
Syntax
createGroq(options?: GroqProviderSettings): GroqProviderParameters
| Name | Type | Description |
|---|---|---|
options | GroqProviderSettings | Optional provider settings, including apiKey and baseURL. |
Returns
A Groq provider instance that can create language models from model identifiers.
Key points
- Import createGroq from the Groq provider package when the default provider instance is not sufficient.
- Call createGroq with a configuration object to create a customized provider.
- Without a custom endpoint, requests use Groq's OpenAI-compatible API prefix.
- The configured API key is sent in the Authorization header.
Common mistakes
- If apiKey is omitted, the provider uses GROQ_API_KEY from the environment.
- Set baseURL when requests must go through a proxy or another compatible endpoint.
Example: Create a configured Groq provider
import { createGroq } from '@ai-sdk/groq';
const groq = createGroq({
apiKey: process.env.GROQ_API_KEY,
});
const model = groq('llama-3.1-8b-instant');Creates a configured Groq provider and selects a model from it.
When to use it
Custom API endpoint
Route Groq-compatible requests through a proxy or alternate endpoint.
Multiple configurations
Create separate provider instances with different settings.
Related
groq@ai-sdk/groqbaseURLapiKey