AnthropicVertexChatGenerator

Generate text using Claude models through the Anthropic Vertex AI API.

Key Features

• Supports Claude models (for example, Claude 3.5 Sonnet, Claude 3 Opus) through the Vertex AI API endpoint.
• Returns responses in the ChatMessage format for chat-based pipelines.
• Supports tool and function calling with configurable tool behavior.
• Configurable generation parameters including temperature, top_p, max tokens, and stop sequences.
• Supports streaming for real-time token delivery.

Configuration

Authentication

Connect Haystack Platform to Anthropic on the Integrations page first. For details, see Use Anthropic Models.

You also need a GCP project with Vertex AI enabled. Create secrets for REGION and PROJECT_ID. For details on secrets, see Add a secret.

1. Drag the AnthropicVertexChatGenerator 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, claude-sonnet-4@20250514).
4. Go to the Advanced tab to configure the region, project ID, generation kwargs, streaming callback, tools, timeout, and max retries.

Connections

AnthropicVertexChatGenerator accepts a list of ChatMessage instances as input (messages). It outputs a list of ChatMessage replies (replies).

Typically, you connect ChatPromptBuilder to the messages input to build the prompt. Connect the replies output to OutputAdapter and then to DeepsetAnswerBuilder to format the final answer.

Usage Example

Using the Component in a Pipeline

This is an example of a RAG pipeline with AnthropicVertexChatGenerator:

components:
  bm25_retriever: # Selects the most similar documents from the document store
    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: 'Standard-Index-English'
          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: 20 # The number of results to return
      fuzziness: 0

  query_embedder:
    type: deepset_cloud_custom_nodes.embedders.nvidia.text_embedder.DeepsetNvidiaTextEmbedder
    init_parameters:
      normalize_embeddings: true
      model: intfloat/e5-base-v2

  embedding_retriever: # Selects the most similar documents from the document store
    type: haystack_integrations.components.retrievers.opensearch.embedding_retriever.OpenSearchEmbeddingRetriever
    init_parameters:
      document_store:
        type: haystack_integrations.document_stores.opensearch.document_store.OpenSearchDocumentStore
        init_parameters:
          hosts:
          index: 'Standard-Index-English'
          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: 20 # The number of results to return

  document_joiner:
    type: haystack.components.joiners.document_joiner.DocumentJoiner
    init_parameters:
      join_mode: concatenate

  ranker:
    type: deepset_cloud_custom_nodes.rankers.nvidia.ranker.DeepsetNvidiaRanker
    init_parameters:
      model: intfloat/simlm-msmarco-reranker
      top_k: 8

  meta_field_grouping_ranker:
    type: haystack.components.rankers.meta_field_grouping_ranker.MetaFieldGroupingRanker
    init_parameters:
      group_by: file_id
      subgroup_by:
      sort_docs_by: split_id

  answer_builder:
    type: deepset_cloud_custom_nodes.augmenters.deepset_answer_builder.DeepsetAnswerBuilder
    init_parameters:
      reference_pattern: acm

  ChatPromptBuilder:
    type: haystack.components.builders.chat_prompt_builder.ChatPromptBuilder
    init_parameters:
      template:
      - _content:
        - text: "You are a helpful assistant answering the user's questions based on the provided documents.\nIf the answer is not in the documents, rely on the web_search tool to find information.\nDo not use your own knowledge.\n"
        _role: system
      - _content:
        - text: "Provided documents:\n{% for document in documents %}\nDocument [{{ loop.index }}] :\n{{ document.content }}\n{% endfor %}\n\nQuestion: {{ query }}\n"
        _role: user
      required_variables:
      variables:
  OutputAdapter:
    type: haystack.components.converters.output_adapter.OutputAdapter
    init_parameters:
      template: '{{ replies[0] }}'
      output_type: List[str]
      custom_filters:
      unsafe: false

  AnthropicVertexChatGenerator:
    type: haystack_integrations.components.generators.anthropic.chat.vertex_chat_generator.AnthropicVertexChatGenerator
    init_parameters:
      region:
      project_id:
      model: claude-3-5-sonnet@20240620
      streaming_callback:
      generation_kwargs:
      ignore_tools_thinking_messages: true
      tools:

connections:  # Defines how the components are connected
- sender: bm25_retriever.documents
  receiver: document_joiner.documents
- sender: query_embedder.embedding
  receiver: embedding_retriever.query_embedding
- sender: embedding_retriever.documents
  receiver: document_joiner.documents
- sender: document_joiner.documents
  receiver: ranker.documents
- sender: ranker.documents
  receiver: meta_field_grouping_ranker.documents
- sender: meta_field_grouping_ranker.documents
  receiver: answer_builder.documents
- sender: meta_field_grouping_ranker.documents
  receiver: ChatPromptBuilder.documents
- sender: OutputAdapter.output
  receiver: answer_builder.replies
- sender: AnthropicVertexChatGenerator.replies
  receiver: OutputAdapter.replies
- sender: ChatPromptBuilder.prompt
  receiver: AnthropicVertexChatGenerator.messages

inputs:  # Define the inputs for your pipeline
  query:  # These components will receive the query as input
  - "bm25_retriever.query"
  - "query_embedder.text"
  - "ranker.query"
  - "answer_builder.query"
  - "ChatPromptBuilder.query"
  filters:  # These components will receive a potential query filter as input
  - "bm25_retriever.filters"
  - "embedding_retriever.filters"

outputs:  # Defines the output of your pipeline
  documents: "meta_field_grouping_ranker.documents"  # The output of the pipeline is the retrieved documents
  answers: "answer_builder.answers"  # The output of the pipeline is the generated answers

max_runs_per_component: 100

metadata: {}

Parameters

Inputs

Parameter	Type	Default	Description
messages	List[ChatMessage]		A list of ChatMessage instances representing the input messages.
streaming_callback	Optional[StreamingCallbackT]	None	A callback function that is called when a new token is received from the stream.
generation_kwargs	Optional[Dict[str, Any]]	None	Optional arguments to pass to the Anthropic generation endpoint.
tools	Optional[Union[List[Tool], Toolset]]	None	A list of Tool objects or a Toolset that the model can use. Each tool should have a unique name. If set, it will override the tools parameter set during component initialization.

Outputs

Parameter	Type	Default	Description
replies	List[ChatMessage]		The responses from the model.

Init Parameters

These are the parameters you can configure in Pipeline Builder:

Parameter	Type	Default	Description
region	str		The region where the Anthropic model is deployed. Defaults to "us-central1".
project_id	str		The GCP project ID where the Anthropic model is deployed.
model	str	claude-sonnet-4@20250514	The name of the model to use.
streaming_callback	Optional[Callable[[StreamingChunk], None]]	None	A callback function that is called when a new token is received from the stream. The callback function accepts StreamingChunk as an argument.
generation_kwargs	Optional[Dict[str, Any]]	None	Other parameters to use for the model. These parameters are all sent directly to the AnthropicVertex endpoint. See Anthropic documentation for more details. Supported generation_kwargs parameters are: - system: The system message to be passed to the model. - max_tokens: The maximum number of tokens to generate. - metadata: A dictionary of metadata to be passed to the model. - stop_sequences: A list of strings that the model should stop generating at. - temperature: The temperature to use for sampling. - top_p: The top_p value to use for nucleus sampling. - top_k: The top_k value to use for top-k sampling. - extra_headers: A dictionary of extra headers to be passed to the model (that is for beta features).
ignore_tools_thinking_messages	bool	True	Anthropic's approach to tools (function calling) resolution involves a "chain of thought" messages before returning the actual function names and parameters in a message. If ignore_tools_thinking_messages is True, the generator will drop so-called thinking messages when tool use is detected. See the Anthropic tools for more details.
tools	Optional[List[Tool]]	None	A list of Tool objects that the model can use. Each tool should have a unique name.
timeout	Optional[float]	None	Timeout for Anthropic client calls. If not set, it defaults to the default set by the Anthropic client.
max_retries	Optional[int]	None	Maximum number of retries to attempt for failed requests. If not set, it defaults to the default set by the Anthropic client.

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