# Create an AI knowledge base assistant with Ollama and PGVector: n8n workflow

> Started by an incoming Telegram message, and works across Telegram. It puts a local model behind a retrieval step, so answers are drawn from documents you...

Source: https://millionrounds.com/workflow/create-an-ai-knowledge-base-assistant-with-ollama-and-pgvect (Millionrounds; every figure on the page links to its filing or article)

[Automation library](https://millionrounds.com/workflows) · [Chat assistants](https://millionrounds.com/workflows?category=Chat+assistants#library)

# Create an AI knowledge base assistant with Ollama and PGVector

Started by an incoming Telegram message, and works across Telegram. It puts a local model behind a retrieval step, so answers are drawn from documents you supply rather than from the model alone. 22 steps, 2 branches: large, and worth reading before you trust it with anything live.

Large Calls a model: a local model Retrieval

[Download free](https://millionrounds.com/login?next=%2Fapp%2Fworkflow%2Fcreate-an-ai-knowledge-base-assistant-with-ollama-and-pgvect)

**22 steps** Large: read it before trusting it with anything live

**2** Branches: places the flow splits

**0** Loops or waits

**1** Service it connects to

## The flow

Starts it 2 Model 8 Logic 4 Service 8

The steps in the order they run, from what starts it.

- 1 **On form submission** Form trigger *Starts it*
- 2 **Embeddings Ollama** Embeddings Ollama *Model*
- 3 **Default Data Loader** Document Default Data Loader *Model*
- 4 **Ollama Chat Model** Lm Chat Ollama *Model*
- 5 **Embeddings Ollama1** Embeddings Ollama *Model*
- 6 **Telegram Trigger** Telegram trigger *Starts it*
- 7 **If** If *Logic*
- 8 **Postgres PGVector Store** Vector Store PGVector *Model*
- 9 **Question and Answer Chain** Chain Retrieval Qa *Model*
- 10 **Postgres PGVector Store1** Vector Store PGVector *Model*
- 11 **Form1** Form *Service*
- 12 **Form** Form *Service*
- 13 **No Operation, do nothing** No Op *Logic*
- 14 **Send a text message** Telegram *Service*
- 15 **Vector Store Retriever** Retriever Vector Store *Model*
- 16 **Switch** Switch *Logic*
- 17 **No Operation, do nothing1** No Op *Logic*
- 18 **Extract from File** Extract From File *Service*
- 19 **Extract from File1** Extract From File *Service*
- 20 **Extract from File2** Extract From File *Service*
- 21 **Extract from File5** Extract From File *Service*
- 22 **Extract from File4** Extract From File *Service*

## How it's built

- **Starts on:** An incoming Telegram message
- **Steps:** 22, with 2 branches
- **Size:** Large: read it before trusting it with anything live
- **Connects to:** Telegram
- **Model:** a local model
- **Retrieval:** Answers are drawn from documents you supply, not from the model alone

## Questions

What does it need to run?

An n8n account, plus your own Telegram account, and a key for a local model. You connect your own accounts after importing it; none are included.

What starts it?

It starts on an incoming Telegram message.

Does it use AI?

Yes. It calls a local model. Its answers draw on documents you give it, not on the model alone. Model calls are billed by the provider, not by n8n.

How long does it take to set up?

Take your time. At 22 steps it is large: read it before trusting it with anything live.

How do I import it?

Copy the workflow, open a new workflow in n8n and paste. The steps appear on the canvas, ready for your own accounts.

**Before you run it**

Nothing here has been tested by us. It is somebody's working automation, read and described from its own steps. Treat the connections it wants as a list of accounts it will touch, and read the 22 steps before you point it at anything that matters.

## Run it in n8n

- **Create your n8n account** Free to start. n8n is where the workflow runs. [Create an n8n account ↗](https://millionrounds.com/n8n)
- **Download the workflow** Free with a Millionrounds account. [Download free](https://millionrounds.com/login?next=%2Fapp%2Fworkflow%2Fcreate-an-ai-knowledge-base-assistant-with-ollama-and-pgvect)
- **Paste it into n8n** Open a new workflow in n8n and paste. Then connect your own accounts.

Stored with the author's traces removed: canvas notes, instance identifiers, saved credential names and any live document they had left behind.

## Connects to

Each opens the other workflows that use it.

[Telegram](https://millionrounds.com/workflows?service=Telegram#library)

## Workflows like it

Also chat assistants, with the same services where possible, closest in size first.

- [**AI agent - RAG Telegram chatbot n8n with OpenAI** 22 steps · calls a model · Google Drive, Supabase vectors](https://millionrounds.com/workflow/ai-agent-rag-telegram-chatbot-n8n-with-openai)
- [**AI-powered Telegram task assistant with Notion** 22 steps · calls a model · Notion, Telegram](https://millionrounds.com/workflow/ai-powered-telegram-task-assistant-with-notion)
- [**Audit landing page conversions with DeepSeek and Google Sheets** 22 steps · calls a model · Gmail, Google Sheets](https://millionrounds.com/workflow/audit-landing-page-conversions-with-deepseek-and-google-shee)
- [**Empower your AI chatbot with OpenAI and Gmail** 22 steps · calls a model · Gmail, Google Docs](https://millionrounds.com/workflow/empower-your-ai-chatbot-with-openai-and-gmail)
- [**Generate AI camera moves with Seedance** 22 steps · calls a model · ClickUp, Google Drive](https://millionrounds.com/workflow/generate-ai-camera-moves-with-seedance)
- [**Get a daily cybersecurity news digest on Telegram and Slack** 22 steps · calls a model · Gmail, Google Sheets](https://millionrounds.com/workflow/get-a-daily-cybersecurity-news-digest-on-telegram-and-slack)