n8n-nodes-grok-reasoning

An n8n community node that adds an xAI Grok Chat Model with a Reasoning Effort control.

It is a faithful clone of n8n's built-in xAI Grok Chat Model node (@n8n/n8n-nodes-langchain.lmChatXAiGrok) — same connection style, same options — with one addition: the reasoning_effort parameter, so you can trade off thinking depth against latency directly from the node.

Connect it to an AI Agent or Basic LLM Chain node as the Chat Model sub-node.

Features

• Works as a Chat Model sub-node for AI Agent / AI Chain nodes.
• Dynamically loads available Grok models from the xAI API.
• Reasoning Effort dropdown: none · low · medium · high.
• Token usage shown in the n8n Logs panel (prompt / completion / total), just like the built-in Grok node — plus the reasoningTokens count when the model reports it, so you can confirm a higher effort really is thinking more.
• Standard options: temperature, top-p, max tokens, frequency/presence penalty, JSON response format, timeout, retries.

Reasoning Effort

| Value | Behaviour | | -------- | --------- | | none | Disables reasoning entirely; no thinking tokens. Fastest. | | low | Some reasoning tokens, but still fast. Good for general agent work. (default) | | medium | More thinking for less latency-sensitive applications. | | high | More reasoning tokens for deeper thinking. Best for hard problems. |

The value is sent to the xAI API as the reasoning_effort body parameter. Only reasoning-capable Grok models honour it (e.g. grok-4 and newer). When reasoning is active (anything other than none), the xAI API ignores temperature, frequency_penalty and presence_penalty, so the node omits them automatically to avoid request errors.

Installation

In n8n: Settings → Community Nodes → Install, then enter:

n8n-nodes-grok-reasoning

Or install manually:

npm install n8n-nodes-grok-reasoning

Credentials

Create an xAI Grok API credential:

• API Key — your key from https://console.x.ai/.
• Base URL — defaults to https://api.x.ai/v1.

Usage

1. Add an AI Agent (or Basic LLM Chain) node.
2. Add the xAI Grok Chat Model (Reasoning Effort) node and connect it to the agent's Chat Model input.
3. Pick a model, set Reasoning Effort, and configure any other options.

License

MIT