Wrapper around Groq API for large language models fine-tuned for chat

Groq API is compatible to the OpenAI API with some limitations. View the full API ref at:

To use, you should have the GROQ_API_KEY environment variable set.

const model = new ChatGroq({
temperature: 0.9,
apiKey: process.env.GROQ_API_KEY,

const response = await model.invoke([new HumanMessage("Hello there!")]);

Hierarchy (view full)



client: Groq
model: string = "mixtral-8x7b-32768"
modelName: string = "mixtral-8x7b-32768"
streaming: boolean = false
temperature: number = 0.7
maxTokens?: number
stop?: string[]
stopSequences?: string[]


  • Parameters

    • request: ChatCompletionCreateParamsStreaming
    • Optionaloptions: any

    Returns Promise<AsyncIterable<ChatCompletionChunk>>

  • Parameters

    • request: ChatCompletionCreateParamsNonStreaming
    • Optionaloptions: any

    Returns Promise<ChatCompletion>