Anthropic claude-3-7-sonnet thinking mode doesnt work with Basic LLM Chain

Describe the problem/error/question

apparently you can’t use the claude 3.7 with enabled thinking in basic llm nodes.

An error occurs when executing the Anthropic Chat Model node (type: @n8n/n8n-nodes-langchain.lmChatAnthropic, version 1.3) configured to use the model “claude-3-7-sonnet-20250219” with the “thinking” option enabled. The attached workflow JSON reproduces the issue.

What is the error message (if any)?

{
“nodes”: [
{
“parameters”: {
“model”: {
“__rl”: true,
“mode”: “list”,
“value”: “claude-3-7-sonnet-20250219”,
“cachedResultName”: “Claude 3.7 Sonnet”
},
“options”: {
“thinking”: true
}
},
“type”: “@n8n/n8n-nodes-langchain.lmChatAnthropic”,
“typeVersion”: 1.3,
“position”: [
540,
220
],
“id”: “6728d52d-8b1f-4ea4-acb9-1666bb461812”,
“name”: “Anthropic Chat Model”,
“credentials”: {
“anthropicApi”: {
“id”: “6IpHJEoyxWoeAvSy”,
“name”: “hacheng1 Anthropic account”
}
}
}
],
“connections”: {
“Anthropic Chat Model”: {
“ai_languageModel”: [

]
}
},
“pinData”: {},
“meta”: {
“templateCredsSetupCompleted”: true,
“instanceId”: “f91b0b942f68abc36946915b0c9cffe661493bd31ec6576b820cb389080a2205”
}
}

Please share your workflow

{
  "nodes": [
    {
      "parameters": {},
      "type": "n8n-nodes-base.manualTrigger",
      "typeVersion": 1,
      "position": [
        0,
        0
      ],
      "id": "54566109-d246-48c1-9f51-491481c9fcb1",
      "name": "When clicking ‘Test workflow’"
    },
    {
      "parameters": {
        "promptType": "define",
        "text": "give me the top 3 marketing strategies for dentists"
      },
      "type": "@n8n/n8n-nodes-langchain.chainLlm",
      "typeVersion": 1.6,
      "position": [
        440,
        0
      ],
      "id": "c2b84b22-9f13-43bc-b3e3-becb4ee116ae",
      "name": "Basic LLM Chain"
    },
    {
      "parameters": {
        "model": {
          "__rl": true,
          "mode": "list",
          "value": "claude-3-7-sonnet-20250219",
          "cachedResultName": "Claude 3.7 Sonnet"
        },
        "options": {
          "thinking": true
        }
      },
      "type": "@n8n/n8n-nodes-langchain.lmChatAnthropic",
      "typeVersion": 1.3,
      "position": [
        540,
        220
      ],
      "id": "6728d52d-8b1f-4ea4-acb9-1666bb461812",
      "name": "Anthropic Chat Model",
      "credentials": {
        "anthropicApi": {
          "id": "6IpHJEoyxWoeAvSy",
          "name": "hacheng1 Anthropic account"
        }
      }
    }
  ],
  "connections": {
    "When clicking ‘Test workflow’": {
      "main": [
        [
          {
            "node": "Basic LLM Chain",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Anthropic Chat Model": {
      "ai_languageModel": [
        [
          {
            "node": "Basic LLM Chain",
            "type": "ai_languageModel",
            "index": 0
          }
        ]
      ]
    }
  },
  "pinData": {},
  "meta": {
    "templateCredsSetupCompleted": true,
    "instanceId": "f91b0b942f68abc36946915b0c9cffe661493bd31ec6576b820cb389080a2205"
  }
}
(Select the nodes on your canvas and use the keyboard shortcuts CMD+C/CTRL+C and CMD+V/CTRL+V to copy and paste the workflow.)

Share the output returned by the last node

Information on your n8n setup

  • n8n version: 1.84.3
  • Database (default: SQLite):
  • n8n EXECUTIONS_PROCESS setting (default: own, main):
  • Running n8n via (Docker, npm, n8n cloud, desktop app): selfhosted
  • Operating system: mac os

Hey @Hanson_Cheng ,

Were you able to make it work with any other type of AI node?
Maybe the AI Agent node, for example.

This is possibly and issue to report, but please let me know more details about it.

By the way, there has been a recent fix to the Basic LLM node. I don’t know if it will help you with your issue, but it’s worth trying to upgrade to 1.85.4 to see if it behaves differently.

It works with AI Agent, but not with basic llm node. It is a bug that needs to be fixed

I was able to reproduce the issue here and have reported the bug.
You can track the progress on this issue here

If this answers your question, please remember to mark it as a solution.

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.