Safety Settings in Google geminı Chat Model doesn't work

[xconkhi9x]

Bug Description

When I texted Gemini, though adjusting the Safety Category to Block_none all. They are still censored and do not send feedback.
I tested the same question on AI Studio and Gemini completely answered.

To Reproduce

1. Send Message
2. Wait Gemini responded
3. Error

Expected behavior

{
"errorMessage": "[GoogleGenerativeAI Error]: Candidate was blocked due to SAFETY",
"errorDetails": {},
"n8nDetails": {
"time": "06:51:07 19/1/2025",
"n8nVersion": "1.74.3 (Self Hosted)",
"binaryDataMode": "default",
"cause": {
"response": {
"candidates": [
{
"finishReason": "SAFETY",
"safetyRatings": [
{
"category": "HARM_CATEGORY_HATE_SPEECH",
"probability": "LOW"
},
{
"category": "HARM_CATEGORY_DANGEROUS_CONTENT",
"probability": "NEGLIGIBLE"
},
{
"category": "HARM_CATEGORY_HARASSMENT",
"probability": "MEDIUM",
"blocked": true
},
{
"category": "HARM_CATEGORY_SEXUALLY_EXPLICIT",
"probability": "NEGLIGIBLE"
}
]
}
],
"usageMetadata": {
"promptTokenCount": 1326,
"candidatesTokenCount": 65,
"totalTokenCount": 1391
},
"modelVersion": "gemini-2.0-flash-exp"
}
}
}
}

Operating System

Ubuntu linux 22.04

n8n Version

1.74.3

Node.js Version

20.18.1

Database

SQLite (default)

Execution mode

main (default)

[Joffcom]

Hey @xconkhi9x,

We have created an internal ticket to look into this which we will be tracking as "N8N-8158"

[burivuhster]

@xconkhi9x , can you please help us to reproduce this issue by sharing the workflow JSON?

[xconkhi9x]

Main_2.json

[jeanpaul]

@xconkhi9x thank you for sharing the workflow. Unfortunately, I can't reproduce the error. Can you send us the input that triggered the error?