How can I use "Claude 3 Sonnet" in GoogleVertexAI?

[kyawthetkt]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

import { NextResponse } from 'next/server';
import { GoogleVertexAI } from "@langchain/community/llms/googlevertexai";
import { getChatModel } from '@/app/util/chat-model';
import { getGoogleAuthCredentails } from '@/app/util/helper';
import { GOOGLE_CLOUD_INFO } from '@/app/util/configs';

export async function GET(request) {
    try {
        const model = new GoogleVertexAI({
            temperature: 0.2,
            // endpoint: GOOGLE_CLOUD_INFO.ENDPOINT,
            // location: "us-central1s",//"asia-southeast1",
            authOptions: {
                credentials: await getGoogleAuthCredentails("any")
            },
            model: "claude-3-sonnet@20240229"
        });

        const data = await model.invoke("Send me a recipe for banana bread.");

        return NextResponse.json({ data });

    } catch (error) {
        return NextResponse.json({ error: error.message });
    }
}

Description

I am trying to use Claude 3 Sonnet in Langchain GoogleVertexAI?

System Info

Platform - linux
Node Version - 18.12.1
NPM - 9.8.1

[dosubot[bot]]

Hey there @kyawthetkt! 🚀 Great to see you diving into the depths of LangChain with Claude 3 Sonnet. Let's make some digital poetry together! How's everything going on your end since our last chat?

Based on the context provided, it seems like you're on the right track with using the GoogleVertexAI class to interact with the "Claude 3 Sonnet" model. However, there's a slight adjustment needed in your code.

The model parameter in the GoogleVertexAI constructor should be a string that represents the model ID, not the model name. The model ID for "Claude 3 Sonnet" should be provided by Google Vertex AI.

Here's how you can adjust your code:

const model = new GoogleVertexAI({
    temperature: 0.2,
    authOptions: {
        credentials: await getGoogleAuthCredentails("any")
    },
    model: "model-id-for-claude-3-sonnet" // replace this with the actual model ID
});

Then, you can use the invoke method to generate a response from the model:

const data = await model.invoke("Send me a recipe for banana bread.");

Please replace "model-id-for-claude-3-sonnet" with the actual model ID for "Claude 3 Sonnet". You can usually find this in the Google Vertex AI console or in the model details page.

If you're still having trouble, please provide more details about the error message you're seeing or the behavior you're experiencing. This will help me provide a more accurate solution.

For more information, you can refer to the LangChain JS GoogleVertexAI example and the LangChain JS GoogleVertexAI source code.

Sources

• examples/src/llms/googlevertexai.ts
• libs/langchain-community/src/llms/googlevertexai/index.ts
• examples/src/llms/googlevertexai-code-bison_legacy.ts
• examples/src/models/chat/integration_googlevertexai-examples.ts

About Dosu

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.