Mindket
English
English简体中文
Sign inSign up
StoreMCPSkillsCalculatorLearnPriceBlog
Discover
Research312Writing59Business95Legal13Finance & accounting33Marketing & SEO175Developer1,422Productivity391Education81Companions14
Help and support
Language
English简体中文

Ollama local model call that writes a welcome message

by Mindketn8n workflowDeveloper
Newversion 1 · updated
Nodes2Apps1Like

Overview

  1. When clicking "Execute Workflow"
  2. Ollama Request
Ollama local model call that writes a welcome message, picture 1
‹2 / 2›

An n8n workflow with two nodes that asks a model running on your own computer through Ollama to write a short welcome message. It is the smallest way to test that n8n can reach Ollama.

Ollama local model call that writes a welcome message is an n8n workflow in the Developer category, listed by Mindket. It is free. This is version 1, last updated on . Download it on this page or copy its JSON, then import it into an n8n canvas. Nothing runs until you activate it in n8n.

At a glance

Runs in
  • Your own n8n on the same computer as Ollama: leave the address as it is
  • n8n in Docker: use the Ollama container’s name or your computer’s address
  • n8n Cloud: only works if Ollama has an address it can reach
  • Ollama: any model you have pulled can replace phi3

Platform

n8nWorkflow automation, in n8n’s cloud or on your own server
Connects 1 app

How it works

Trigger

Manual run

It starts when you press Test workflow, and there is no form:

  • No input: the model name and the prompt are set in the Ollama Request node

Actions

  1. An HTTP request posts the model phi3 and a prompt to Ollama at localhost on port 11434
  2. Ollama writes a brief greeting for someone who likes AI automation
  3. The answer comes back as one JSON object, with the text in its response field

Result

One item with the welcome message in its response field, plus the model name and timing figures

Integrations

What you need before you start.

Ollama

Local language model

Runs phi3 on your own computer and answers the prompt.

NeedsOllama running with the phi3 model pulled

Workflow check

Checked by Mindket on 13 Nov 2026

Steps
2 steps, starting by hand
AI
No step calls an AI model
Sign-ins
No sign-in
API keys
No API key is left in the file
Step types
Mindket’s checker reads all 2 of its step types

Read from the workflow’s file, not from a run.

About

What it is

This n8n workflow is the smallest test of a local model. A manual trigger starts it, and one HTTP request sends a prompt to Ollama on your own computer. Ollama answers with a single JSON object, and its response field holds a brief greeting aimed at people who like automation. There are two nodes and no account to sign in to.

Who it suits

Developers and tinkerers who run Ollama and want to check that n8n can reach it, or who want a base to build a local AI workflow on. Skip it if you need a chat, memory or tools, because this is one request with one answer.

Who uses it, and how
IndustryWho uses itHow they use itExample
SoftwareDevelopersCheck the link between n8n and Ollama before building moreA developer runs it once and reads the welcome message to see that the model answers.
Privacy-minded teamsIT staffTry a model that keeps prompts on the company networkAn IT admin points the address at an Ollama server in the office.
EducationTeachersShow students a language model that runs offlineA teacher runs it live and then edits the prompt together with the class.
AutomationFreelancersStart a local text step for client workflowsA freelancer copies the node into a larger flow that writes short texts.

You need little: Ollama running, the phi3 model pulled and an n8n that can reach it.

Where to use it

An n8n on the same computer or network as Ollama. n8n Cloud cannot reach localhost on your own computer.

  • Your own n8n on the same computer as Ollama: leave the address as it is
  • n8n in Docker: use the Ollama container’s name or your computer’s address
  • n8n Cloud: only works if Ollama has an address it can reach
  • Ollama: any model you have pulled can replace phi3

Setup

Download saves a zip with workflow.json and Mindket’s setup notes in your downloads folder. You only install Ollama and pull one model.

What you need first
NeedDetail
n8nAny version, on a machine that can reach Ollama
OllamaInstalled and running, on port 11434 by default
Modelphi3, pulled with ollama pull phi3
  1. Install Ollama and start it. In a terminal, run ollama pull phi3.
  2. In n8n, create a workflow, open its menu, choose Import from File and pick workflow.json.
  3. Open Ollama Request. Leave the address if n8n and Ollama share a machine, and change localhost if they do not.
  4. Press Test workflow and wait for the first answer.

How to use it

Press Test workflow. The output item holds the text in its response field. Change the prompt in the request body to ask for something else, and raise the timeout if the first answer is slow.

What to change for a different result
To changeWhere
The modelThe model value in the body of Ollama Request
The questionThe prompt value in the same body
The addressThe url of Ollama Request
The waitThe timeout in the options of Ollama Request

Example

A developer has just installed Ollama and wants to see it work.

The developer presses Test workflow. After a short wait the output shows a welcome message in the response field, with the model name and the timing figures beside it. If it fails, the address or the pulled model is the first thing to check.

For newcomers and experts

If you are new. n8n draws a workflow as boxes joined by lines, and each box, a node, does one job. Ollama is a program that runs language models on your own computer. This workflow sends it one question and shows the answer. Try it before you build anything bigger.

For experts. The body sets stream to false, so Ollama returns one object, and a stream would arrive as many small objects instead.

Where to look closer
NodeWhat it doesWhat to know
The manual triggerStarts the run by handIt has no settings
Ollama RequestPosts the model, the prompt and the stream setting to /api/generateIt times out after 60 seconds

Good to know

Nothing here is retired, the timeout can be short for a first run and the data stays on your side.

  • Retirement: Ollama’s API page, read on 12 November 2026, still documents /api/generate with the stream setting, and Ollama’s library still lists phi3. Nothing in this workflow is retired
  • Copy: the workflow is the original’s, unchanged, because it had no credential and no private address
  • Timeout: a slow computer or a model that is not loaded yet can run past 60 seconds. Raise it in the options of Ollama Request
  • Address: n8n’s Ollama page, read on 12 November 2026, suggests 127.0.0.1 when localhost fails, and the container’s name for two Docker containers on one network
  • Keys: the request sends none, so it suits an Ollama on your own computer or network
  • Billing: no vendor bills this, and each request costs only the power of your computer
  • Your data: the prompt and the reply go only to the Ollama address you set
  • Credentials: the JSON holds none

README and files

README

Ollama local model call that writes a welcome message

An n8n workflow with two nodes. A manual trigger starts it, and an HTTP request asks a model that runs on your own
computer through Ollama to write a short welcome message. Ollama answers with one JSON object, and its response field
holds the text.

What is inside

  • workflow.json: the workflow as n8n exports it, 2 nodes and no sticky note. It holds no credential.
  • README.md: these notes.

What we changed from the original copy

  • Nothing in the workflow. The original copy had no credential, no private address and no sticky note, so Mindket's copy
    is the same two nodes with the same settings.

Before you start

  • Ollama installed and running on the computer that runs n8n.
  • The phi3 model pulled into Ollama.
  • An n8n that can reach Ollama. n8n Cloud cannot reach localhost on your own computer.

Set it up

  1. Install Ollama and start it. In a terminal, run ollama pull phi3 to download the model.
  2. In n8n, create a workflow, open its menu and choose Import from File. Pick workflow.json.
  3. Open Ollama Request. If n8n and Ollama run on the same machine, leave the address as it is. If not, replace
    localhost with the address where Ollama answers. n8n's Ollama page, read on 12 November 2026, suggests 127.0.0.1 when
    localhost fails. For two Docker containers on one network, it suggests the container's name, such as
    http://ollama:11434.
  4. Press Test workflow. The first request can take a while, because Ollama loads the model before it answers.

How a run goes

  1. The manual trigger starts the run.
  2. Ollama Request posts to /api/generate with the model phi3, the prompt and "stream": false. That last setting
    makes Ollama answer with one JSON object instead of a stream. The request times out after 60 seconds.
  3. The output is one item. Its response field holds the welcome message, and it also carries the model name, a
    timestamp, a done flag and timing figures.

Change the look

  • The model: the model value in the body of Ollama Request. Any model you have pulled works.
  • The question: the prompt value in the same body.
  • The address: the url of Ollama Request.
  • The wait: the timeout in its options, which is 60000 milliseconds.
  • More settings: add system or options to the body. Ollama's API page lists them.

Good to know

  • Ollama's API page, read on the same day, still documents /api/generate with the stream setting, and Ollama's
    library still lists phi3. Nothing in this workflow is retired.
  • A slow computer or a model that is not loaded yet can run past the 60 second timeout. Raise it in the options if the
    node fails on the first try.
  • The reply is plain text in response. A later node can read it with {{ $json.response }}.
  • The request sends no key, so it suits an Ollama on your own computer or network.
  • The prompt and the reply go only to the Ollama address you set. Each request costs only the power of your computer.
Files
NOTICE.md  560 B
README.md  3 KB
workflow.json  1 KB

3 files, 5 KB, downloaded as a zip.

What’s new

  • Version 1 · 2026-11-12

Reviews

No ratings yet

Write a review

No reviews yet.

More from Mindket

See all from Mindket

anaksor-ocr

Sokheng · Kit

Reads Khmer and English text from documents, receipts and photos of signs, using a detector for text lines and a recogniser for each line. Suits developers who…

New

Qwen2.5-7B-Instruct-GGUF by Qwen

Qwen · Kit

Qwen2.5-7B-Instruct as GGUF files in eight quantization sizes, to chat with locally through llama.cpp. For people who want a capable 7.6B model on their own co…

New

ACE-Step 1.5 XL SFT

ACE-Step · Kit

Generates music from text with a 4B diffusion decoder, 50 steps and guidance control, aiming for higher audio quality than the base. For creators with a GPU of…

New

Similar in Developer

See all in Developer

Claude Code Skills & Plugins

alirezarezvani · Skill

Adds 411 ready-made skills for Claude Code and other coding agents, grouped by role from engineering and product to marketing, finance and compliance. Installs…

New29.8k

buzz

block · Kit

Self-hostable workspace where people and AI agents share the same rooms, built on a Nostr relay that logs every action as a signed event. For teams that want t…

New37.3k

plannotator

backnotprop · Kit

Opens your AI coding agent's plans, specs and code changes in a browser so you can annotate or comment on them and send the feedback straight back to the agent.

New10.5k