Gemini Pro 3 - Thought Signature Error

Questions
1

Describe the problem/error/question

Using Gemini Pro 3 in AI agent node (with tools connected also) results in a missing thought signature error.

What is the error message (if any)?

{
“errorMessage”: “Bad request - please check your parameters”,
“errorDescription”: “[GoogleGenerativeAI Error]: Error fetching from https://generativelanguage.googleapis.com/v1beta/models/gemini-3-pro-preview:generateContent: [400 Bad Request] Function call is missing a thought_signature in functionCall parts. This is required for tools to work correctly, and missing thought_signature may lead to degraded model performance. Additional data, function call default_api:Get_Projects_List , position 2. please refer to Thought Signatures  |  Gemini API  |  Google AI for Developers for more details.”,
“errorDetails”: {},
“n8nDetails”: {
“time”: “11/18/2025, 3:17:23 PM”,
“n8nVersion”: “1.120.3 (Cloud)”,
“binaryDataMode”: “filesystem”
}
}

Please share your workflow

Share the output returned by the last node

Bad request - please check your parameters
[GoogleGenerativeAI Error]: Error fetching from https://generativelanguage.googleapis.com/v1beta/models/gemini-3-pro-preview:generateContent: [400 Bad Request] Function call is missing a thought_signature in functionCall parts. This is required for tools to work correctly, and missing thought_signature may lead to degraded model performance. Additional data, function call default_api:Get_Projects_List , position 2. please refer to Thought Signatures  |  Gemini API  |  Google AI for Developers for more details.

Information on your n8n setup

  • n8n version: 1.120.3 (cloud)
  • Database (default: SQLite):
  • n8n EXECUTIONS_PROCESS setting (default: own, main):
  • Running n8n via (Docker, npm, n8n cloud, desktop app): n8n cloud
  • Operating system:
7 Likes
2

Looks like a change needed on n8n’s implememtation:

When using Gemini 3 Pro, you must pass back thought signatures during function calling, otherwise you will get a validation error (4xx status code).

(from the references documentation)

2 Likes
3

Does anybody know how this could be implemented now, maybe using code?

Or do we have to wait for n8n to add support for thought signatures in their next version?

1 Like
4

ran into this issue as well.

5

Same here. I set the Agent Node to 2.2 because of a failure with version 3. now this pops up. With Gemini 2.5 pro it works.

6

I am having the same issue. I was using grok-4.1-fast:free from openai but then after 2 days of use it became unfree, so I said lets try gemini 3 pro preview. This is the error message I got immediately.

7

I had to ditch gemini 3 pro preview as well because of it. There’s activity in n8n’s github for this: chore: Update langchain by schrothbn · Pull Request #22500 · n8n-io/n8n · GitHub

Just not clear when it will be released in the cloud or elsewhere.

9

Same problema here! Really wanna know when this issue will be resolved?