EveryLocalAI

Posted on Jun 14

Build Visual AI Agent Pipelines with Langflow and Ollama

#beginners #tutorial #ai #python

Prototype and deploy multi-agent and RAG applications with a visual drag-and-drop interface - all running locally with your own models.

Langflow is an open-source visual framework for building AI applications. Connect it to Ollama for local inference, and you get a powerful environment for designing agent architectures, RAG pipelines, and chatbot workflows without writing code.

What You Need

A GPU with 12GB+ VRAM (or CPU-only for prototyping)
Docker or Python 3.10+
About 15 minutes

Architecture

Component	Role
Langflow	Visual drag-and-drop flow builder and API server
Ollama	Serves local LLM models
Qwen3 14B	Default model - fits 12GB at Q4

Setup

Option A: Docker (Recommended)

Save this as docker-compose.yml:

services:
  ollama:
    image: ollama/ollama:latest
    container_name: ollama
    volumes:
      - ollama:/root/.ollama
    ports:
      - "11434:11434"
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities: [gpu]
    restart: unless-stopped

  langflow:
    image: langflowai/langflow:latest
    container_name: langflow
    depends_on:
      - ollama
    ports:
      - "7860:7860"
    volumes:
      - langflow_data:/app/langflow
    environment:
      - LANGFLOW_AUTO_LOGIN=true
    restart: unless-stopped

volumes:
  ollama:
  langflow_data:

Launch it:

docker compose up -d
docker exec ollama ollama pull qwen3:14b

Open http://localhost:7860 to access Langflow.

Option B: pip Install

pip install langflow
langflow run
# In another terminal:
ollama pull qwen3:14b

Open http://localhost:7860.

Connect Langflow to Ollama

In the Langflow canvas, add:

Ollama Chat Model component - Base URL: http://ollama:11434 (Docker) or http://localhost:11434 (pip)
Select model: qwen3:14b
Connect to a Prompt node and Chat Output for a basic chatbot

What You Can Build

RAG Chatbot

Drag in: File > Ollama Embeddings > Vector Store (Chroma) > Ollama Chat Model > Chat Output. Upload a PDF, ask questions - answers come from your documents.

Multi-Agent Research System

Add an Agent node with a Web Search Tool + Ollama, add a second Agent for summarization. One agent gathers info, the other condenses it.

Document Processing Pipeline

Combine File Loader > Splitter > Ollama Embeddings > Vector Store. Add Ollama Chat Model with custom prompts for Q&A over your documents.

Cost vs Cloud

	Local Langflow + Ollama	Langflow Cloud + OpenAI
Monthly	$0	$50-200+
Hardware	~$300-600 once	$0
Data privacy	Stays on your machine	Sent to cloud
AI calls	Unlimited, free	Per-token billing

Full guide with detailed troubleshooting and alternatives: https://everylocalai.com/stack/langflow-ollama-rag-agent

DEV Community