Google Vertex AI

Since Camel 4.17

Only producer is supported

This component integrates with Google Cloud Vertex AI, supporting both Google’s native models (Gemini, Imagen) and partner models (Claude, Llama, Mistral) through the rawPredict API.

Prerequisites

  • A Google Cloud Platform account with Vertex AI API enabled

  • A service account with appropriate permissions, or Application Default Credentials configured

For setup instructions, see Vertex AI Documentation.

URI Format

google-vertexai:projectId:location:modelId[?options]

  • projectId - Your GCP project ID

  • location - Region where the model is available (e.g., us-central1, us-east5)

  • modelId - Model identifier (e.g., gemini-2.0-flash, claude-sonnet-4@20250514)

Configuring Options

Configuring Options

Camel components are configured on two separate levels:

  • component level

  • endpoint level

Configuring Component Options

At the component level, you set general and shared configurations that are, then, inherited by the endpoints. It is the highest configuration level.

For example, a component may have security settings, credentials for authentication, urls for network connection and so forth.

Some components only have a few options, and others may have many. Because components typically have pre-configured defaults that are commonly used, then you may often only need to configure a few options on a component; or none at all.

You can configure components using:

  • the Component DSL.

  • in a configuration file (application.properties, *.yaml files, etc).

  • directly in the Java code.

Configuring Endpoint Options

You usually spend more time setting up endpoints because they have many options. These options help you customize what you want the endpoint to do. The options are also categorized into whether the endpoint is used as a consumer (from), as a producer (to), or both.

Configuring endpoints is most often done directly in the endpoint URI as path and query parameters. You can also use the Endpoint DSL and DataFormat DSL as a type safe way of configuring endpoints and data formats in Java.

A good practice when configuring options is to use Property Placeholders.

Property placeholders provide a few benefits:

  • They help prevent using hardcoded urls, port numbers, sensitive information, and other settings.

  • They allow externalizing the configuration from the code.

  • They help the code to become more flexible and reusable.

The following two sections list all the options, firstly for the component followed by the endpoint.

Component Options

The Google Vertex AI component supports 18 options, which are listed below.

Name Description Default Type

serviceAccountKey (common)

The Service account key that can be used as credentials for the Vertex AI client. It can be loaded by default from classpath, but you can prefix with classpath:, file:, or http: to load the resource from different systems.

String

anthropicVersion (producer)

Anthropic API version for Claude models. Required when publisher is 'anthropic'.

vertex-2023-10-16

String

candidateCount (producer)

Number of candidate responses to generate.

1

Integer

configuration (producer)

The component configuration.

GoogleVertexAIConfiguration

jsonMode (producer)

Whether to use JSON request/response format.

false

boolean

lazyStartProducer (producer)

Whether the producer should be started lazy (on the first message). By starting lazy you can use this to allow CamelContext and routes to startup in situations where a producer may otherwise fail during starting and cause the route to fail being started. By deferring this startup to be lazy then the startup failure can be handled during routing messages via Camel’s routing error handlers. Beware that when the first message is processed then creating and starting the producer may take a little time and prolong the total processing time of the processing.

false

boolean

maxOutputTokens (producer)

Maximum number of output tokens.

1024

Integer

operation (producer)

Set the operation for the producer.

Enum values:

  • generateText

  • generateChat

  • generateChatStreaming

  • generateImage

  • generateEmbeddings

  • generateCode

  • generateMultimodal

  • rawPredict

  • streamRawPredict

GoogleVertexAIOperations

publisher (producer)

Publisher name for partner models (e.g., anthropic, meta, mistralai). Required for rawPredict operations.

String

streamOutputMode (producer)

Streaming output mode: complete (default) or chunks.

Enum values:

  • complete

  • chunks

complete

String

temperature (producer)

Temperature parameter for generation (0.0-1.0).

0.7

Float

topK (producer)

Top-K parameter for generation.

40

Integer

topP (producer)

Top-P parameter for nucleus sampling.

0.95

Float

autowiredEnabled (advanced)

Whether autowiring is enabled. This is used for automatic autowiring options (the option must be marked as autowired) by looking up in the registry to find if there is a single instance of matching type, which then gets configured on the component. This can be used for automatic configuring JDBC data sources, JMS connection factories, AWS Clients, etc.

true

boolean

client (advanced)

Autowired The Google GenAI client for Vertex AI.

Client

predictionServiceClient (advanced)

Autowired The Google Cloud AI Platform Prediction Service client for rawPredict operations.

PredictionServiceClient

healthCheckConsumerEnabled (health)

Used for enabling or disabling all consumer based health checks from this component.

true

boolean

healthCheckProducerEnabled (health)

Used for enabling or disabling all producer based health checks from this component. Notice: Camel has by default disabled all producer based health-checks. You can turn on producer checks globally by setting camel.health.producersEnabled=true.

true

boolean

Endpoint Options

The Google Vertex AI endpoint is configured using URI syntax:

google-vertexai:projectId:location:modelId

With the following path and query parameters:

Path Parameters (3 parameters)

Name Description Default Type

projectId (common)

Required Google Cloud Project ID.

String

location (common)

Required Google Cloud location/region (e.g., us-central1).

String

modelId (common)

Required Model ID to use for predictions.

String

Query Parameters (14 parameters)

Name Description Default Type

serviceAccountKey (common)

The Service account key that can be used as credentials for the Vertex AI client. It can be loaded by default from classpath, but you can prefix with classpath:, file:, or http: to load the resource from different systems.

String

anthropicVersion (producer)

Anthropic API version for Claude models. Required when publisher is 'anthropic'.

vertex-2023-10-16

String

candidateCount (producer)

Number of candidate responses to generate.

1

Integer

jsonMode (producer)

Whether to use JSON request/response format.

false

boolean

maxOutputTokens (producer)

Maximum number of output tokens.

1024

Integer

operation (producer)

Set the operation for the producer.

Enum values:

  • generateText

  • generateChat

  • generateChatStreaming

  • generateImage

  • generateEmbeddings

  • generateCode

  • generateMultimodal

  • rawPredict

  • streamRawPredict

GoogleVertexAIOperations

publisher (producer)

Publisher name for partner models (e.g., anthropic, meta, mistralai). Required for rawPredict operations.

String

streamOutputMode (producer)

Streaming output mode: complete (default) or chunks.

Enum values:

  • complete

  • chunks

complete

String

temperature (producer)

Temperature parameter for generation (0.0-1.0).

0.7

Float

topK (producer)

Top-K parameter for generation.

40

Integer

topP (producer)

Top-P parameter for nucleus sampling.

0.95

Float

lazyStartProducer (producer (advanced))

Whether the producer should be started lazy (on the first message). By starting lazy you can use this to allow CamelContext and routes to startup in situations where a producer may otherwise fail during starting and cause the route to fail being started. By deferring this startup to be lazy then the startup failure can be handled during routing messages via Camel’s routing error handlers. Beware that when the first message is processed then creating and starting the producer may take a little time and prolong the total processing time of the processing.

false

boolean

client (advanced)

Autowired The Google GenAI client for Vertex AI.

Client

predictionServiceClient (advanced)

Autowired The Google Cloud AI Platform Prediction Service client for rawPredict operations.

PredictionServiceClient

Message Headers

The Google Vertex AI component supports 24 message header(s), which is/are listed below:

Name Description Default Type

CamelGoogleVertexAIOperation (producer)

Constant: OPERATION

The operation to perform.

Enum values:

  • generateText

  • generateChat

  • generateChatStreaming

  • generateImage

  • generateEmbeddings

  • generateCode

  • generateMultimodal

  • rawPredict

  • streamRawPredict

GoogleVertexAIOperations

CamelGoogleVertexAIModelId (producer)

Constant: MODEL_ID

The model ID to use for generation.

String

CamelGoogleVertexAIProjectId (producer)

Constant: PROJECT_ID

The project ID to use for the request.

String

CamelGoogleVertexAILocation (producer)

Constant: LOCATION

The location/region to use for the request.

String

CamelGoogleVertexAITemperature (producer)

Constant: TEMPERATURE

The temperature parameter for generation (0.0-1.0).

Float

CamelGoogleVertexAITopP (producer)

Constant: TOP_P

The top-p parameter for generation.

Float

CamelGoogleVertexAITopK (producer)

Constant: TOP_K

The top-k parameter for generation.

Integer

CamelGoogleVertexAIMaxOutputTokens (producer)

Constant: MAX_OUTPUT_TOKENS

The maximum number of output tokens.

Integer

CamelGoogleVertexAIcandidateCount (producer)

Constant: CANDIDATE_COUNT

The number of candidate responses to generate.

Integer

CamelGoogleVertexAIStreamOutputMode (producer)

Constant: STREAM_OUTPUT_MODE

The streaming output mode (complete or chunks).

String

CamelGoogleVertexAIPrompt (producer)

Constant: PROMPT

The prompt text for text generation.

String

CamelGoogleVertexAIChatMessages (producer)

Constant: CHAT_MESSAGES

The chat messages for chat generation.

List

CamelGoogleVertexAISystemInstruction (producer)

Constant: SYSTEM_INSTRUCTION

The system instruction for the model.

String

CamelGoogleVertexAISafetySettings (producer)

Constant: SAFETY_SETTINGS

The safety settings for content filtering.

List

CamelGoogleVertexAIFinishReason (producer)

Constant: FINISH_REASON

The finish reason from the response.

String

CamelGoogleVertexAIPromptTokenCount (producer)

Constant: PROMPT_TOKEN_COUNT

The number of tokens in the prompt.

Integer

CamelGoogleVertexAICandidatesTokenCount (producer)

Constant: CANDIDATES_TOKEN_COUNT

The number of tokens in the response.

Integer

CamelGoogleVertexAITotalTokenCount (producer)

Constant: TOTAL_TOKEN_COUNT

The total token count (prompt response).

Integer

CamelGoogleVertexAISafetyRatings (producer)

Constant: SAFETY_RATINGS

The safety ratings from the response.

List

CamelGoogleVertexAIContentBlocked (producer)

Constant: CONTENT_BLOCKED

Whether the content was blocked by safety filters.

Boolean

CamelGoogleVertexAIChunkCount (producer)

Constant: STREAMING_CHUNK_COUNT

The number of chunks received in streaming response.

Integer

CamelGoogleVertexAIPublisher (producer)

Constant: PUBLISHER

Publisher name for partner models (e.g., anthropic, meta, mistralai).

String

CamelGoogleVertexAIRawResponse (producer)

Constant: RAW_RESPONSE

The raw JSON response from rawPredict operation.

String

CamelGoogleVertexAIAnthropicVersion (producer)

Constant: ANTHROPIC_VERSION

Anthropic API version for Claude models.

String

Authentication

The component supports two authentication methods:

  1. Service Account Key File - Set the serviceAccountKey parameter to a file path:

    serviceAccountKey=file:/path/to/service-account.json

  2. Application Default Credentials - If no key is provided, the component uses ADC automatically.

Operations

Google Models

Use these operations with Gemini, Imagen, and other Google models:

Operation Description

generateText

Generate text from a prompt

generateChat

Generate chat responses

generateCode

Generate code from a description

Partner Models

Use rawPredict for Claude, Llama, and Mistral models:

Operation Description

rawPredict

Send requests to partner models via the Prediction Service API

Partner models require the publisher parameter: anthropic, meta, or mistralai.

Examples

Text Generation with Gemini

from("direct:prompt")
    .setBody(constant("Explain quantum computing briefly"))
    .to("google-vertexai:my-project:us-central1:gemini-2.0-flash")
    .log("${body}");

Using a Service Account

from("direct:prompt")
    .to("google-vertexai:my-project:us-central1:gemini-2.0-flash"
        + "?serviceAccountKey=file:/path/to/key.json")
    .log("${body}");

Calling Claude with rawPredict

from("direct:claude")
    .setBody(constant("What is Apache Camel?"))
    .to("google-vertexai:my-project:us-east5:claude-sonnet-4@20250514"
        + "?operation=rawPredict&publisher=anthropic")
    .log("${body}");

The component automatically wraps simple text prompts in the required Anthropic format.

Claude with Custom Request

For more control, pass a JSON or Map body:

from("direct:claude")
    .process(exchange -> {
        String json = """
            {
                "anthropic_version": "vertex-2023-10-16",
                "max_tokens": 512,
                "messages": [{"role": "user", "content": "Write a haiku"}]
            }
            """;
        exchange.getMessage().setBody(json);
    })
    .to("google-vertexai:my-project:us-east5:claude-sonnet-4@20250514"
        + "?operation=rawPredict&publisher=anthropic")
    .log("${body}");

Supported Models

Google Models

Use with generateText, generateChat, or generateCode:

  • Gemini: gemini-2.5-pro, gemini-2.5-flash, gemini-2.0-flash, gemini-1.5-pro, gemini-1.5-flash

  • Imagen: imagen-4.0-generate-preview-05-20, imagen-3.0-generate-002

  • Gemma: gemma-3-27b-it, gemma-2-27b-it, gemma-2-9b-it

  • Embeddings: text-embedding-005, text-embedding-004

Partner Models

Use with rawPredict and the appropriate publisher:

Publisher Models Regions

anthropic

claude-sonnet-4@20250514, claude-opus-4@20250514, claude-3-5-haiku@20241022

us-east5, europe-west1

meta

llama-4-maverick-17b-128e-instruct-maas, llama-3.3-70b-instruct-maas

us-central1

mistralai

mistral-medium-3, mistral-small-2503, codestral-2

us-central1, europe-west4

See VertexAIModels for all available model identifiers.

Dependencies

Add the following dependency:

<dependency>
    <groupId>org.apache.camel</groupId>
    <artifactId>camel-google-vertexai</artifactId>
    <version>x.x.x</version>
</dependency>

Spring Boot Auto-Configuration

When using google-vertexai with Spring Boot make sure to use the following Maven dependency to have support for auto configuration:

<dependency>
  <groupId>org.apache.camel.springboot</groupId>
  <artifactId>camel-google-vertexai-starter</artifactId>
  <version>x.x.x</version>
  <!-- use the same version as your Camel core version -->
</dependency>

The component supports 19 options, which are listed below.

Name Description Default Type

camel.component.google-vertexai.anthropic-version

Anthropic API version for Claude models. Required when publisher is 'anthropic'.

vertex-2023-10-16

String

camel.component.google-vertexai.autowired-enabled

Whether autowiring is enabled. This is used for automatic autowiring options (the option must be marked as autowired) by looking up in the registry to find if there is a single instance of matching type, which then gets configured on the component. This can be used for automatic configuring JDBC data sources, JMS connection factories, AWS Clients, etc.

true

Boolean

camel.component.google-vertexai.candidate-count

Number of candidate responses to generate.

1

Integer

camel.component.google-vertexai.client

The Google GenAI client for Vertex AI. The option is a com.google.genai.Client type.

Client

camel.component.google-vertexai.configuration

The component configuration. The option is a org.apache.camel.component.google.vertexai.GoogleVertexAIConfiguration type.

GoogleVertexAIConfiguration

camel.component.google-vertexai.enabled

Whether to enable auto configuration of the google-vertexai component. This is enabled by default.

Boolean

camel.component.google-vertexai.health-check-consumer-enabled

Used for enabling or disabling all consumer based health checks from this component.

true

Boolean

camel.component.google-vertexai.health-check-producer-enabled

Used for enabling or disabling all producer based health checks from this component. Notice: Camel has by default disabled all producer based health-checks. You can turn on producer checks globally by setting camel.health.producersEnabled=true.

true

Boolean

camel.component.google-vertexai.json-mode

Whether to use JSON request/response format.

false

Boolean

camel.component.google-vertexai.lazy-start-producer

Whether the producer should be started lazy (on the first message). By starting lazy you can use this to allow CamelContext and routes to startup in situations where a producer may otherwise fail during starting and cause the route to fail being started. By deferring this startup to be lazy then the startup failure can be handled during routing messages via Camel’s routing error handlers. Beware that when the first message is processed then creating and starting the producer may take a little time and prolong the total processing time of the processing.

false

Boolean

camel.component.google-vertexai.max-output-tokens

Maximum number of output tokens.

1024

Integer

camel.component.google-vertexai.operation

Set the operation for the producer.

GoogleVertexAIOperations

camel.component.google-vertexai.prediction-service-client

The Google Cloud AI Platform Prediction Service client for rawPredict operations. The option is a com.google.cloud.aiplatform.v1.PredictionServiceClient type.

PredictionServiceClient

camel.component.google-vertexai.publisher

Publisher name for partner models (e.g., anthropic, meta, mistralai). Required for rawPredict operations.

String

camel.component.google-vertexai.service-account-key

The Service account key that can be used as credentials for the Vertex AI client. It can be loaded by default from classpath, but you can prefix with classpath:, file:, or http: to load the resource from different systems.

String

camel.component.google-vertexai.stream-output-mode

Streaming output mode: complete (default) or chunks.

complete

String

camel.component.google-vertexai.temperature

Temperature parameter for generation (0.0-1.0).

Float

camel.component.google-vertexai.top-k

Top-K parameter for generation.

40

Integer

camel.component.google-vertexai.top-p

Top-P parameter for nucleus sampling.

Float
Back to top