VertexAITextGenerator

Generate text using Google Vertex AI generative models.

Key Features

• Generates text responses using Vertex AI models (text-bison, text-unicorn, text-bison-32k).
• Authenticates using Google Cloud Application Default Credentials.
• Returns generated replies along with safety attributes and citations.

Configuration

Authentication

This component authenticates using Google Cloud Application Default Credentials (ADCs). For more information, see the official Google documentation.

Create secrets for GCP_PROJECT_ID and optionally GCP_DEFAULT_REGION. For detailed instructions on creating secrets, see Create Secrets.

1. Drag the VertexAITextGenerator component onto the canvas from the Component Library.
2. Click the component to open the configuration panel.
3. On the General tab:
   1. Enter the model name (for example, text-bison).
4. Go to the Advanced tab to configure the project ID, location, and additional model kwargs.

Connections

VertexAITextGenerator accepts a prompt string as input. It outputs a list of generated replies (replies), safety attribute scores (safety_attributes), and citations (citations).

Typically, you connect PromptBuilder to the prompt input and AnswerBuilder to the replies output.

Usage Example

This query pipeline uses VertexAITextGenerator to generate text responses:

components:
  bm25_retriever:
    type: haystack_integrations.components.retrievers.opensearch.bm25_retriever.OpenSearchBM25Retriever
    init_parameters:
      document_store:
        type: haystack_integrations.document_stores.opensearch.document_store.OpenSearchDocumentStore
        init_parameters:
          hosts:
          index: 'default'
          max_chunk_bytes: 104857600
          embedding_dim: 768
          return_embedding: false
          method:
          mappings:
          settings:
          create_index: true
          http_auth:
          use_ssl:
          verify_certs:
          timeout:
      top_k: 10
      fuzziness: 0

  PromptBuilder:
    type: haystack.components.builders.prompt_builder.PromptBuilder
    init_parameters:
      template: |
        Given the following information, answer the question.

        Context:
        {% for document in documents %}
        {{ document.content }}
        {% endfor %}

        Question: {{ query }}
      required_variables:
      variables:

  VertexAITextGenerator:
    type: haystack_integrations.components.generators.google_vertex.text_generator.VertexAITextGenerator
    init_parameters:
      project_id:
      model: text-bison
      location:

  AnswerBuilder:
    type: haystack.components.builders.answer_builder.AnswerBuilder
    init_parameters:
      pattern:
      reference_pattern:

connections:
  - sender: bm25_retriever.documents
    receiver: PromptBuilder.documents
  - sender: PromptBuilder.prompt
    receiver: VertexAITextGenerator.prompt
  - sender: VertexAITextGenerator.replies
    receiver: AnswerBuilder.replies
  - sender: bm25_retriever.documents
    receiver: AnswerBuilder.documents

inputs:
  query:
    - bm25_retriever.query
    - PromptBuilder.query
    - AnswerBuilder.query

outputs:
  answers: AnswerBuilder.answers

max_runs_per_component: 100

metadata: {}

Parameters

Inputs

Parameter	Type	Default	Description
prompt	str		The prompt to use for text generation.

Outputs

Parameter	Type	Default	Description
replies	List[str]		A list of generated replies.
safety_attributes	Dict[str, float]		Safety scores for each answer.
citations	List[Dict[str, Any]]		Citations for each answer.

Init Parameters

These are the parameters you can configure in Pipeline Builder:

Parameter	Type	Default	Description
project_id	Optional[str]	None	ID of the GCP project to use. By default, it is set during Google Cloud authentication.
model	str	text-bison	Name of the model to use.
location	Optional[str]	None	The default location to use when making API calls. If not set, uses us-central-1.
kwargs	Any		Additional keyword arguments to pass to the model. See the TextGenerationModel.predict() documentation.

Run Method Parameters

These are the parameters you can configure for the component's run() method. This means you can pass these parameters at query time through the API, in Playground, or when running a job. For details, see Modify Pipeline Parameters at Query Time.

Parameter	Type	Default	Description
prompt	str		The prompt to use for text generation.