Hi n8n team, I’m Jay (Nguyen Thieu Toan), an Verified n8n Creator, and I’d like to raise a concern that is starting to affect quite a few users and clients who are trying to adopt DeepSeek inside n8n for production AI Agent workflows. Right now, whenever DeepSeek Chat Model is used with AI Agent +…

Hi @achamm Chat Model for AI Agent dont have this option. :frowning:

@nguyenthieutoan u can use community nodes or diff agent for now i suggest making github issue so n8n dev team can pick up

The root cause is how n8n/LangChain appends messages to the history during tool execution loops; DeepSeek is strict about receiving its own generated reasoning_content back in subsequent API calls, or it breaks with a 400 error.

Thanks @sergeys - that’s a precise explanation of the root cause. So the fix would need to happen at the LangChain message history layer, where n8n strips or preserves reasoning_content between tool calls. That’s not something we can patch at the node level without changes to how the AI Agent struct…

Will n8n release an official fix for DeepSeek tool calls soon?

Questions

achamm June 14, 2026, 1:55pm 2

Can you just use HTTP Requests?

Topic		Replies	Views
DeepSeek Chat Model got issues with reasoning_content Questions core , node	9	150	June 17, 2026
DeepSeek AI Agent fails with 400 "reasoning_content must be passed back" when using tools Questions ai , deepseek	5	137	June 16, 2026
I’ve fixed the reasoning_content error on DeepSeek Chat Model Built with n8n core , node , n8n-node , langchain , ai , deepseek	0	34	June 16, 2026
Deepseek LLM infinite loop on tool usage Questions community-node , openai	6	1499	February 10, 2025
How to connect an http request or DeepSeek v3 as a chat model? Questions node , http-request , n8n-node	30	6597	March 2, 2025

Will n8n release an official fix for DeepSeek tool calls soon?

Related topics