Vertext Ai Chat model thrown Exception Expect a map object but found array when using gemini chat model

[kiditz]

It seems that the server returns a JSON structure that is not being correctly interpreted as a map object by the client code.

---

Environment

• Spring AI version: 1.0.0-M7
• Java version: 21
• Vector Store: None
• Other: Using MCP server as backend.

---

Steps to reproduce

• Use Vertext AI Gemini model 1.5flash

private ChatModel getVertexChatModel(String name, double temperature) {
		VertexAI vertexApi = new VertexAI(projectId, location);
		return VertexAiGeminiChatModel.builder()
				.vertexAI(vertexApi)
				.defaultOptions(VertexAiGeminiChatOptions.builder()
						.model(name)
						.temperature(temperature)
						.build())
				.build();
	}

• Use mcp server return products

@mcp.tool(description="List products from Shopify")
def list_products(limit: int = 10) -> dict:
    client = get_shopify_client()
    products = client.list_products(params={"limit": limit})
    return {"products": products}

• Ask Chat Client

list 2 products

at org.springframework.ai.vertexai.gemini.VertexAiGeminiChatModel.jsonToStruct(VertexAiGeminiChatModel.java:408) ~[spring-ai-vertex-ai-gemini-1.0.0-M7.jar:1.0.0-M7]
	at org.springframework.ai.vertexai.gemini.VertexAiGeminiChatModel.lambda$messageToGeminiParts$1(VertexAiGeminiChatModel.java:368) ~[spring-ai-vertex-ai-gemini-1.0.0-M7.jar:1.0.0-M7]
	at java.base/java.util.stream.ReferencePipeline$3$1.accept(ReferencePipeline.java:197) ~[na:na]
	at java.base/java.util.ArrayList$ArrayListSpliterator.forEachRemaining(ArrayList.java:1708) ~[na:na]
	at java.base/java.util.stream.AbstractPipeline.copyInto(AbstractPipeline.java:509) ~[na:na]
	at java.base/java.util.stream.AbstractPipeline.wrapAndCopyInto(AbstractPipeline.java:499) ~[na:na]
	at java.base/java.util.stream.AbstractPipeline.evaluate(AbstractPipeline.java:575) ~[na:na]
	at java.base/java.util.stream.AbstractPipeline.evaluateToArrayNode(AbstractPipeline.java:260) ~[na:na]
	at java.base/java.util.stream.ReferencePipeline.toArray(ReferencePipeline.java:616) ~[na:na]
	at java.base/java.util.stream.ReferencePipeline.toArray(ReferencePipeline.java:622) ~[na:na]
	at java.base/java.util.stream.ReferencePipeline.toList(ReferencePipeline.java:627) ~[na:na]
	at org.springframework.ai.vertexai.gemini.VertexAiGeminiChatModel.messageToGeminiParts(VertexAiGeminiChatModel.java:371) ~[spring-ai-vertex-ai-gemini-1.0.0-M7.jar:1.0.0-M7]
	at org.springframework.ai.vertexai.gemini.VertexAiGeminiChatModel.lambda$toGeminiContent$20(VertexAiGeminiChatModel.java:781) ~[spring-ai-vertex-ai-gemini-1.0.0-M7.jar:1.0.0-M7]
	at java.base/java.util.stream.ReferencePipeline$3$1.accept(ReferencePipeline.java:197) ~[na:na]
	at java.base/java.util.AbstractList$RandomAccessSpliterator.forEachRemaining(AbstractList.java:722) ~[na:na]
	at java.base/java.util.stream.AbstractPipeline.copyInto(AbstractPipeline.java:509) ~[na:na]
	at java.base/java.util.stream.AbstractPipeline.wrapAndCopyInto(AbstractPipeline.java:499) ~[na:na]
	at java.base/java.util.stream.AbstractPipeline.evaluate(AbstractPipeline.java:575) ~[na:na]
	at java.base/java.util.stream.AbstractPipeline.evaluateToArrayNode(AbstractPipeline.java:260) ~[na:na]
	at java.base/java.util.stream.ReferencePipeline.toArray(ReferencePipeline.java:616) ~[na:na]
	at java.base/java.util.stream.ReferencePipeline.toArray(ReferencePipeline.java:622) ~[na:na]
	at java.base/java.util.stream.ReferencePipeline.toList(ReferencePipeline.java:627) ~[na:na]
	at org.springframework.ai.vertexai.gemini.VertexAiGeminiChatModel.toGeminiContent(VertexAiGeminiChatModel.java:783) ~[spring-ai-vertex-ai-gemini-1.0.0-M7.jar:1.0.0-M7]
	at org.springframework.ai.vertexai.gemini.VertexAiGeminiChatModel.createGeminiRequest(VertexAiGeminiChatModel.java:742) ~[spring-ai-vertex-ai-gemini-1.0.0-M7.jar:1.0.0-M7]
	at org.springframework.ai.vertexai.gemini.VertexAiGeminiChatModel.lambda$internalStream$12(VertexAiGeminiChatModel.java:556) ~[spring-ai-vertex-ai-gemini-1.0.0-M7.jar:1.0.0-M7]
	at reactor.core.publisher.FluxDeferContextual.subscribe(FluxDeferContextual.java:49) ~[reactor-core-3.7.4.jar:3.7.4]
	at reactor.core.publisher.FluxDefer.subscribe(FluxDefer.java:54) ~[reactor-core-3.7.4.jar:3.7.4]
	at reactor.core.publisher.FluxSubscribeOn$SubscribeOnSubscriber.run(FluxSubscribeOn.java:194) ~[reactor-core-3.7.4.jar:3.7.4]
	at reactor.core.scheduler.BoundedElasticThreadPerTaskScheduler$SchedulerTask.run(BoundedElasticThreadPerTaskScheduler.java:1013) ~[reactor-core-3.7.4.jar:3.7.4]
	at java.base/java.lang.VirtualThread.run(VirtualThread.java:309) ~[na:na]
Caused by: com.google.protobuf.InvalidProtocolBufferException: Expect a map object but found: [{"text":"{\"products\": {\"products\": [{\"id\": \"gid://shopify/Product/8484113744026\", \"handle\": 

The issue is when using vertex ai only, while i change the model to OpenAiChatModel, it's working

[ddobrin]

Could you consider switching the model to the latest released model:
https://ai.google.dev/gemini-api/docs/models#gemini-2.0-flash

Flash 1.5 support expires in September 2025.

[faranahmadk]

Hi team,

I am facing the same error while implementing client-server MCP using Spring AI with Vertex AI Gemini (latest model):

com.google.protobuf.InvalidProtocolBufferException: Expect a map object but found:

Any solutions from your side. Thanks in advance!

[kiditz]

Could you consider switching the model to the latest released model: https://ai.google.dev/gemini-api/docs/models#gemini-2.0-flash

Flash 1.5 support expires in September 2025.

Sorry for late reply i was on mountain hike lately, i have try other gemini such as 2.0 Pro, 2.0 Flash, the result still the same

Caused by: com.google.protobuf.InvalidProtocolBufferException: Expect a map object but found: [{"text":"{\"products\": {\"products\": [{\"id\": 8484507975834, \"title\": \"Example Product\", \"body_html\": null, \"vendor\": \"Toelbot\", \"product_type\": \"\", \"created_at\": \"2025-04-20T18:00:50-04:00\", \"handle\": \"example-product\", \"updated_at\": \"2025-04-22T20:50:59-04:00\", \"published_at\": \"2025-04-20T18:00:50-04:00\", \"template_suffix\": null, \"published_scope\": \"global\", \"tags\": \"\", \"status\": \"active\", \"admin_graphql_api_id\": \"gid://shopify/Product/8484507975834\", \"variants\": [{\"id\": 45685146976410, \"product_id\": 8484507975834, \"title\": \"Default Title\", \"price\": \"0.00\", 

[markpollack]

please try the snapshots and reopen if it doesn't solve your issue.

[faranahmadk]

It is WORKING !!! Thanks alot