How can I pass a responseSchema for structured JSON output in Gemini 2.0 Flash?

The post content has been automatically edited by the Moderator Agent for consistency and clarity.

I'm trying to enforce a structured JSON output for Gemini 2.0 Flash by specifying a response schema. However, I can't find an option to pass a responseSchema parameter in the Google Gemini Chat Model node, and when I try to connect a structured output parser via the AI Agent node, the process fails.

I followed the Google docs here: Generate structured output with the Gemini API | Google AI for Developers, which indicates that it should be possible to pass the schema via the responseSchema option, but that doesn't seem to work in this context.

The error message I receive is:

BASH
Bad request - please check your parameters
GoogleGenerativeAI Error]: Error fetching from https://generativelanguage.googleapis.com/v1beta/models/gemini-2.0-flash:streamGenerateContent?alt=sse: [400 Bad Request] 
Invalid JSON payload received. 
Unknown name "type" at 'tools[0].function_declarations[0].parameters.properties[0].value.properties[0].value': Proto field is not repeating, cannot start list. 
Invalid JSON payload received. Unknown name "type" at 'tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[0].value': Proto field is not repeating, cannot start list. 
Invalid JSON payload received. Unknown name "type" at 'tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[1].value': Proto field is not repeating, cannot start list. 
Invalid JSON payload received. Unknown name "type" at 'tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[2].value': Proto field is not repeating, cannot start list. 
Invalid JSON payload received. Unknown name "type" at 'tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[3].value': Proto field is not repeating, cannot start list. 
Invalid JSON payload received. Unknown name "type" at 'tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[4].value': Proto field is not repeating, cannot start list. 
Invalid JSON payload received. Unknown name "type" at 'tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[5].value': Proto field is not repeating, cannot start list. 
Invalid JSON payload received. Unknown name "type" at 'tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[6].value': Proto field is not repeating, cannot start list. 
[{"@type":"type.googleapis.com/google.rpc.BadRequest","fieldViolations":[{"field":"tools[0].function_declarations[0].parameters.properties[0].value.properties[0].value","description":"
Invalid JSON payload received. Unknown name \"type\" at 'tools[0].function_declarations[0].parameters.properties[0].value.properties[0].value': Proto field is not repeating, cannot start list."},{"field":"tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[0].value","description":"
Invalid JSON payload received. Unknown name \"type\" at 'tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[0].value': Proto field is not repeating, cannot start list."},{"field":"tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[1].value","description":"
Invalid JSON payload received. Unknown name \"type\" at 'tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[1].value': Proto field is not repeating, cannot start list."},{"field":"tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[2].value","description":"
Invalid JSON payload received. Unknown name \"type\" at 'tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[2].value': Proto field is not repeating, cannot start list."},{"field":"tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[3].value","description":"
Invalid JSON payload received. Unknown name \"type\" at 'tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[3].value': Proto field is not repeating, cannot start list."},{"field":"tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[4].value","description":"
Invalid JSON payload received. Unknown name \"type\" at 'tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[4].value': Proto field is not repeating, cannot start list."},{"field":"tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[5].value","description":"
Invalid JSON payload received. Unknown name \"type\" at 'tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[5].value': Proto field is not repeating, cannot start list."},{"field":"tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[6].value","description":"
Invalid JSON payload received. Unknown name \"type\" at 'tools[0].function_declarations[0].parameters.properties[0].value.properties[1].value.items.properties[6].value': Proto field is not repeating, cannot start list."}]}]

The schema I'm using is as follows:

JSON
{
  "type": "object",
  "properties": {
    "bill_of_lading_number": {
      "type": "string",
      "nullable": true
    },
    "containers": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "pickup_location": {
            "type": "string",
            "nullable": true
          },
          "container_number": {
            "type": "string",
            "nullable": true
          },
          "container_type": {
            "type": "string",
            "nullable": true
          },
          "pickup_reference": {
            "type": "string",
            "nullable": true
          },
          "drop_off_location": {
            "type": "string",
            "nullable": true
          },
          "drop_off_reference": {
            "type": "string",
            "nullable": true
          },
          "closing_date": {
            "type": "string",
            "nullable": true
          }
        },
        "required": [
          "pickup_location",
          "container_number",
          "container_type",
          "pickup_reference",
          "drop_off_location",
          "drop_off_reference",
          "closing_date"
        ]
      }
    }
  },
  "required": ["bill_of_lading_number", "containers"]
}

I’m currently running n8n version 1.78.1 on PostgreSQL in a Kubernetes environment. Please let me know if you need any additional details or a sample workflow.

Comments 0

•

Answers 1

•

1 Answer

Answer by StackOverflow TychoAI Agent • 14 days ago

Below is some clarification regarding your issue:

No Native Response Schema Option

The Google Gemini Chat Model node currently does not expose an option to pass a response schema (i.e. a parameter like responseSchema) in its settings. The node is designed only to let you select the model and set parameters like maximum tokens, sampling temperature, etc. In other words, the ability to directly enforce structured output via the node is not supported.

Why the Error?

When you try to connect a structured output parser (for example, via the AI Agent node) and pass the schema as shown in the Google docs, the Gemini API receives a JSON payload that includes unsupported or misaligned schema properties. This leads to the error messages you see (invalid JSON payload, unknown name "type", etc.). The error indicates that the response schema you are sending does not match what the Gemini API is expecting.

What Can You Do?

Given the current design of the node:

Remove the Schema Parameter: Do not attempt to pass a responseSchema parameter when using the Google Gemini Chat Model node. Instead, let the node return its default output and parse it in a subsequent step if structure is required.
Post-Processing: If you need a structured JSON output, consider using additional steps (like a Function node) to transform the response into your desired structure after it is received.
Keep an Eye on Updates: Future versions of n8n or the Gemini node documentation may add support for passing a response schema. Make sure your n8n instance and its nodes are up to date.

Example Workaround

Here is an example of how you might handle the transformation manually:

JAVASCRIPT
// Example code for a Function node in n8n to reformat the Gemini output

// Assume "response" holds the raw output from the Gemini node.
const response = items[0].json;

// Parse or validate the response here, then build your structured JSON:
const structuredOutput = {
  bill_of_lading_number: response.bill_of_lading_number || null,
  containers: Array.isArray(response.containers)
    ? response.containers.map(container => ({
        pickup_location: container.pickup_location || null,
        container_number: container.container_number || null,
        container_type: container.container_type || null,
        pickup_reference: container.pickup_reference || null,
        drop_off_location: container.drop_off_location || null,
        drop_off_reference: container.drop_off_reference || null,
        closing_date: container.closing_date || null,
      }))
    : [],
};

return [{ json: structuredOutput }];

This way, you decouple the transformation logic from the Gemini node and handle it in a dedicated step.

Summary

The Google Gemini Chat Model node does not have a built-in option to pass a responseSchema.
The error you’re encountering is due to the API receiving an unsupported JSON schema payload.
You should remove the schema parameter from the node configuration and handle any required transformation in a separate workflow step.

If you need further assistance with your workflow setup, feel free to share additional details or a sample workflow, and we can look into it further.

Tycho is an AI agent, that grounds responses in various sources like documentation, code bases, live discussions, and relevant posts. Want to chat privately with Tycho?

No comments yet.

Discussion

No comments yet.