5,365,291 AI resources indexedWorking toward the world’s largest AI resource index.

Browse resources 13 sections
PDF

Local document question answering with Ollama AI, Agentic RAG & PGVector

Knowledge & search workflow for n8n with 35 active steps. Starts with Chat Trigger, Webhook, Local File Trigger. Uses Ollama, OpenAI. Integrations include PostgreSQL, Ollama, OpenAI.

By Jadai kongoloWorkflowReference information

What it does

35 active nodes · Starts with Chat Trigger, Webhook, Local File Trigger.

Steps in the template include:

  • Default Data Loader (Default Data Loader)
  • Extract Document Text (Extract from File) — text
  • Postgres Chat Memory (Postgres Chat Memory)
  • Set File ID (Edit Fields (Set))

Nodes and integrations

  • Default Data Loader
  • Extract from File ×4
  • Postgres Chat Memory
  • Edit Fields (Set) ×3
  • Respond to Webhook
  • Chat Trigger
  • Webhook
  • Aggregate
  • Summarize
  • AI Agent
  • Switch
  • Postgres ×7
  • postgres Tool ×3
  • Loop Over Items (Split in Batches)
  • Local File Trigger
  • Read/Write Files from Disk
  • Embeddings Ollama ×2
  • Recursive Character Text Splitter
  • OpenAI Chat Model
  • Postgres PGVector Store ×2
See the author’s step names
  • Default Data Loader (Default Data Loader)
  • Extract Document Text (Extract from File) — text
  • Postgres Chat Memory (Postgres Chat Memory)
  • Set File ID (Edit Fields (Set))
  • Respond to Webhook (Respond to Webhook)
  • Edit Fields (Edit Fields (Set))
  • When chat message received (Chat Trigger)
  • Webhook (Webhook)
  • Extract PDF Text (Extract from File) — pdf
  • Aggregate (Aggregate)
  • Summarize (Summarize)
  • RAG AI Agent (AI Agent)
  • Switch (Switch)
  • Extract from Excel (Extract from File) — xlsx
  • Set Schema (Edit Fields (Set))
  • Extract from CSV (Extract from File)
  • Create Document Metadata Table (Postgres) — executeQuery
  • Create Document Rows Table (for Tabular Data) (Postgres) — executeQuery
  • List Documents (postgres Tool) — select
  • Get File Contents (postgres Tool) — executeQuery
  • Query Document Rows (postgres Tool) — executeQuery
  • Loop Over Items (Loop Over Items (Split in Batches))
  • Insert Document Metadata (Postgres) — upsert
  • Insert Table Rows (Postgres)
  • Update Schema for Document Metadata (Postgres) — upsert
  • Local File Trigger (Local File Trigger)
  • Read/Write Files from Disk (Read/Write Files from Disk)
  • Embeddings Ollama (Embeddings Ollama)
  • Embeddings Ollama1 (Embeddings Ollama)
  • Recursive Character Text Splitter (Recursive Character Text Splitter)
  • Ollama (Change Base URL) (OpenAI Chat Model)
  • Delete Old Doc Records (Postgres) — executeQuery
  • Delete Old Data Records (Postgres) — executeQuery
  • Postgres PGVector Store (Postgres PGVector Store)
  • Postgres PGVector Store1 (Postgres PGVector Store)

Accounts and credentials

  • PostgreSQL
  • Ollama instance
  • OpenAI access

AI models and swaps

Ollama · Embeddings

nomic-embed-text:latest in Embeddings Ollama. Changing the embedding model may require rebuilding the vector index and matching its dimensions.

Ollama · Embeddings

nomic-embed-text:latest in Embeddings Ollama1. Changing the embedding model may require rebuilding the vector index and matching its dimensions.

OpenAI · Chat

qwen2.5:14b-8k in Ollama (Change Base URL). Potentially swappable with a compatible n8n chat model node. Match tool calling before testing.

Inputs and setup options

  • When chat message received: public chat is enabled in the template
  • Webhook: HTTP POST

Resource information

Credit basis
Author listed by n8n
Pricing
Template access shown on n8n
License
Not specified
Last record update
Nov 5, 2026