Chat with local LLMs using n8n and Ollama
Chat with local LLMs using n8n and Ollama This n8n workflow allows you to seamlessly interact with your self-hosted Large Language Models (LLMs) through a user-friendly chat interface. By connecting to Ollama, a powerful tool for managing local LLMs, you can send prompts and receive AI-generated responses directly within n8n. Use cases Private AI Interactions Ideal for scenarios where data privacy and confidentiality are important. Cost-Effective LLM Usage Avoid ongoing cloud API costs by running models on your own hardware. Experimentation & Learning A great way to explore and experiment with different LLMs in a local, controlled environment. Prototyping & Development Build and test AI-powered applications without relying on external services. How it works When chat message received: Captures the user's input from the chat interface. Chat LLM Chain: Sends the input to the Ollama server and receives the AI-generated response. Delivers the LLM's response back to the chat interface. Set up steps Make sure Ollama is installed and running on your machine before executing this workflow. Edit the Ollama address if different from the default.
By Mihai FarcasAI-powered WhatsApp chatbot 🤖📲 for text, voice, images & PDFs with memory 🧠
This workflow is a highly advanced multimodal AI assistant designed to operate through WhatsApp. It can understand and respond to text, images, voice messages, and PDF documents by combining OpenAI models with smart logic to adapt to the content received. --- 🎯 Core Features 📥 1. Automatic Message Type Detection Using the Input type node, the bot detects whether the user has sent: Text Voice messages Images Files (PDF) Other unsupported content 💬 2. Smart Text Message Handling Text messages are processed by an OpenAI GPT-4o-mini agent with a customized system prompt. Replies are concise, accurate, and formatted for mobile readability. 🖼️ 3. Image Analysis & Description Images are downloaded, converted to base64, and analyzed by an image-aware AI model. The output is a rich, structured description, designed for visually impaired users or visual content interpretation. 🎙️ 4. Voice Message Transcription & Reply Audio messages are downloaded and transcribed using OpenAI Whisper. The transcribed text is analyzed and answered by the AI. Optionally, the AI reply can be converted back to voice using OpenAI's text-to-speech, and sent as an audio message. 📄 5. PDF Document Extraction & Summary Only PDFs are allowed (filtered via MIME type). The document’s content is extracted and combined with the user's message. The AI then provides a relevant summary or answer. 🧠 6. Contextual Memory Each user has a personalized session ID with a memory window of 10 interactions. This ensures a more natural and contextual conversation flow. --- How It Works Thisworkflow is designed to handle incoming WhatsApp messages and process different types of inputs (text, audio, images, and PDF documents) using AI-powered analysis. Here’s how it functions: Trigger: The workflow starts with the WhatsApp Trigger node, which listens for incoming messages (text, audio, images, or documents). Input Routing: The Input type (Switch node) checks the message type and routes it to the appropriate processing branch: Text: Directly forwards the message to the AI agent for response generation. Audio: Downloads the audio file, transcribes it using OpenAI, and sends the transcription to the AI agent. Image: Downloads the image, analyzes it with OpenAI’s GPT-4 model, and generates a detailed description. PDF Document: Downloads the file, extracts text, and processes it with the AI agent. Unsupported Formats: Sends an error message if the input is not supported. AI Processing: The AI Agent1 node, powered by OpenAI, processes the input (text, transcribed audio, image description, or PDF content) and generates a response. Response Handling: For audio inputs, the AI’s response is converted back into speech (using OpenAI’s TTS) and sent as a voice message. For other inputs, the response is sent as a text message via WhatsApp. Memory: The Simple Memory node maintains conversation context for follow-up interactions. Setup Steps To deploy this workflow in n8n, follow these steps: Configure WhatsApp API Credentials: Set up WhatsApp Business API credentials (Meta Developer Account). Add the credentials in the WhatsApp Trigger, Get Image/Audio/File URL, and Send Message nodes. Set Up OpenAI Integration: Provide an OpenAI API key in the Analyze Image, Transcribe Audio, Generate Audio Response, and AI Agent1 nodes. Adjust Input Handling (Optional): Modify the Switch node ("Input type") to handle additional message types if needed. Update the "Only PDF File" IF node to support other document formats. Test & Deploy: Activate the workflow and test with different message types (text, audio, image, PDF). Ensure responses are correctly generated and sent back via WhatsApp. --- Need help customizing? Contact me for consulting and support or add me on Linkedin.
By DavideGmail AI Email Manager
Want to check out all my flows, follow me on: https://maxmitcham.substack.com/ https://www.linkedin.com/in/max-mitcham/ Email Manager - Intelligent Gmail Classification This automation flow is designed to automatically monitor incoming Gmail messages, analyze their content and context using AI, and intelligently classify them with appropriate labels for better email organization and prioritization. ⚙️ How It Works (Step-by-Step): 📧 Gmail Monitoring (Trigger) Continuously monitors your Gmail inbox: Polls for new emails every minute Captures all incoming messages automatically Triggers workflow for each new email received 📖 Email Content Extraction Retrieves complete email details: Full email body and headers Sender information and recipient lists Subject line and metadata Existing Gmail labels and categories Email threading information (replies/forwards) 🔍 Email History Analysis AI agent checks relationship context: Searches for previous emails from the same sender Checks sent folder for prior outbound correspondence Determines if this is a first-time contact (cold email) Analyzes conversation thread history 🤖 Intelligent Classification Agent Advanced AI categorization using: Claude Sonnet 4 for sophisticated email analysis Context-aware classification based on email history Content analysis for intent and urgency detection Header analysis for automated vs. human-sent emails 🏷️ Smart Label Assignment Automatically applies appropriate Gmail labels: To Respond: Requires direct action/reply FYI: For awareness, no action needed Notification: Service updates, policy changes Marketing: Promotional content and sales pitches Meeting Update: Calendar-related communications Comment: Document/task feedback 📋 Structured Processing Ensures consistent labeling: Uses structured output parsing for reliability Returns specific Label ID for Gmail integration Applies label automatically to the email Maintains classification accuracy 🛠️ Tools Used: n8n: Workflow automation platform Gmail API: Email monitoring and label management Anthropic Claude: Advanced email content analysis Gmail Tools: Email history checking and search Structured Output Parser: Consistent AI responses 📦 Key Features: Real-time email monitoring and classification Context-aware analysis using email history Intelligent cold vs. warm email detection Multiple classification categories for organization Automatic Gmail label application Header analysis for automated email detection Thread-aware conversation tracking 🚀 Ideal Use Cases: Busy executives managing high email volumes Sales professionals prioritizing prospect communications Support teams organizing customer inquiries Marketing teams filtering promotional content Anyone wanting automated email organization Teams needing consistent email prioritization
By Max MitchamInvoice processor & validator with OCR, AI & Google Sheets
📝 Say goodbye to manual invoice checking! This smart workflow automates your entire invoice processing pipeline using AI, OCR, and Google Sheets. --- ⚙️ What This Workflow Does: 📥 1. Reads an invoice PDF — Select a local PDF invoice from your machine. 🔍 2. Extracts raw text using OCR — Converts scanned or digital PDFs into readable text. 🧠 3. AI Agent processes the text — Transforms messy raw text into clean JSON using natural language understanding. 🧱 4. Structures and refines the JSON — Converts AI output into a structured, usable format. 🔄 5. Splits item-wise data — Extracts individual invoice line items with all details. 🆔 6. Generates unique keys — Creates a unique identifier for each item for tracking. 📊 7. Updates Google Sheet — Adds extracted items to your designated sheet automatically. 📂 8. Fetches master item data — Loads your internal product master to validate against. ✅ 9. Validates item name & cost — Compares extracted items with your official records to verify accuracy. 📌 10. Updates results per item — Marks each item as Valid or Invalid in the sheet based on matching. --- 💼 Use Case: Perfect for businesses, freelancers, or operations teams who receive invoices and want to automate validation, detect billing errors, and log everything seamlessly in Google Sheets — all using the power of AI + n8n. > 🔁 Fast. Accurate. Zero manual work. --- OCR AI Invoices Automation.
By Dhrumil PatelTurn YouTube videos into summaries, transcripts, and visual insights
Who is this for? This workflow is built for anyone who works with YouTube content, whether you're: A learner looking to understand a video’s key points A content creator repurposing video material A YouTube manager looking to update titles, descriptions A social media strategist searching for the most shareable clips Don't just ask questions about what's said. Find out what's going on in a video too. Video Overview: https://www.youtube.com/watch?v=Ovg_KfKxnC8 What problem does this solve? YouTube videos hold valuable insights, but watching and processing them manually takes time. This workflow automates: Quick content extraction: Summarize key ideas without watching full videos Visual analysis: Understand what’s happening beyond spoken words Clip discovery: Identify the best moments for social sharing How the workflow works This n8n-powered automation: Uses Google’s Gemini 1.5 Flash AI for intelligent video analysis Provides multiple content analysis templates tailored to different needs What makes this workflow powerful? The easiest place to start is by requesting a summary or transcript. From there, you can refine the prompts to match your specific use case and the type of video content you’re working with. But what's even more amazing? You can ask questions about what’s happening in the video — and get detailed insights about the people, objects, and scenes. It's jaw-dropping. This workflow is versatile — the actions adapt based on the values set. That means you can use a single workflow to: Extract transcripts Generate an extended YouTube description Write a summary blog post You can also modify the trigger based on how you want to run the workflow — use a webhook, connect it to an event in Airtable, or leave it as-is for on-demand use. The output can then be sent anywhere: Notion, Airtable, CMS platforms, or even just stored for reference. How to set it up Connect your Google API key Paste a YouTube video URL Select an analysis method Run the workflow and get structured results Analysis Templates Basic & Timestamped Transcripts: Extract spoken content Summaries: Get concise takeaways Visual Scene Analysis: Detect objects, settings, and people Clip Finder: Locate shareable moments Actionable Insights: Extract practical information Customization Options Modify templates to fit your needs Connect with external platforms Adjust formatting preferences Advanced Configuration This workflow is designed for use with gemini-1.5-flash. In the future, you can update the flow to work with different models or even modify the HTTP request node to define which API endpoint should be used. It's also been designed so you can use this flow on it's own or add to a new / existing worflow. This workflow helps you get the most out of YouTube content — quickly and efficiently.
By Colleen BradyFrom Google Drive to Instagram, TikTok & YouTube with AI descriptions & Airtable tracking
Description This automation template is designed for content creators, social media managers, and influencers who want to streamline their video publishing workflow. It automatically detects new videos uploaded to a specific Google Drive folder, generates AI-powered descriptions based on video audio content, and simultaneously publishes them across Instagram, TikTok, and YouTube while tracking everything in Airtable. Note: This workflow uses upload-post.com API (free trial no credit card required) for multi-platform video distribution and requires API tokens for each service. The AI-generated descriptions are created using OpenAI's transcription and chat models to analyze video audio content. Who Is This For? Content Creators & Influencers: Automatically publish your videos across all major social platforms without manual work. Social Media Managers: Maintain consistent posting schedules across multiple platforms with AI-generated, platform-optimized descriptions. Marketing Teams: Scale video content distribution with automated workflows that include tracking and status monitoring. Video Producers: Focus on creating content while the system handles the tedious task of multi-platform publishing and description generation. What Problem Does This Workflow Solve? Publishing the same video content across Instagram, TikTok, and YouTube is time-consuming and repetitive. You need to manually upload each video, write unique descriptions, and track publication status. This workflow addresses these challenges by: Automated Video Distribution: Detects new videos in Google Drive and automatically uploads them to all three platforms simultaneously. AI-Powered Content Generation: Uses OpenAI to transcribe video audio and generate engaging, platform-appropriate descriptions automatically. Centralized Tracking: Maintains detailed records in Airtable including upload status, URLs, and metadata for each platform. Error Monitoring: Provides real-time error notifications via Telegram to ensure you're always aware of any issues. How It Works Video Upload Detection: The workflow monitors a specific Google Drive folder for new video uploads using automated triggers. Content Analysis: Downloads the video, extracts audio, and uses OpenAI to transcribe and generate compelling descriptions. Airtable Integration: Creates and updates records to track video metadata, descriptions, and publication status. Multi-Platform Publishing: Simultaneously uploads the video to Instagram, TikTok, and YouTube using the upload-post.com API. Status Tracking: Updates Airtable records with publication status and platform-specific URLs for each successful upload. Setup Google Drive Configuration: Set up the Google Drive trigger to monitor your specific folder Configure OAuth2 credentials for Google Drive access OpenAI Integration: Add your OpenAI API key to enable audio transcription and description generation Airtable Setup: Create an Airtable base with fields for Video Name, Description, Platform Status, URLs, and Upload Date Add your Airtable API token and configure base/table IDs in the "Set Variables" node Upload-Post.com Account: Create an account at upload-post.com to get your API token Configure the token in the HTTP request nodes for each platform Set your user ID in the variables section Platform Accounts: Ensure your Instagram, TikTok, and YouTube accounts are connected to upload-post.com Error Notifications: (Optional) Configure Telegram bot credentials for error notifications Requirements Accounts: Google Drive, OpenAI, Airtable, upload-post.com, Telegram (optional) API Keys & Credentials: Google Drive OAuth2, OpenAI API Key, Airtable API Token, upload-post.com API Token Platform Setup: Instagram, TikTok, and YouTube accounts connected to upload-post.com Transform your video publishing workflow from hours of manual work to a fully automated system that handles everything from content analysis to multi-platform distribution and tracking.
By Juan Carlos Cavero GraciaTurn on a light to a specific color on any update in GitHub repository
This workflow turns a light red when an update is made to a GitHub repository. By default, updates include pull requests, issues, pushes just to name a few. Prerequisites GitHub credentials. Home Assistant credentials. How it works Triggers off on the On any update in repository node. Uses Home Assistant to turn on a light and then configure the light to turn red.
By n8n TeamAutomate Instagram & Facebook posting with Meta Graph API & System User Tokens
This template automates posting to Instagram Business and Facebook Pages using the Meta Graph API. It supports both short-lived and long-lived tokens, with a secure approach using System User tokens for reliable, ongoing automation. Includes detailed guidance for authentication, token refresh logic, and API use. Features: 📸 Publish to Instagram via /media + /media_publish 📘 Post to Facebook Pages via /photos 🔐 Long-lived token support via Meta Business System User ♻️ Token refresh support using staticData in n8n 🧠 In-line sticky note instructions Use Cases: Schedule and publish branded social media content Automate marketing flows with CRM + social sync Empower internal teams or clients to post without manual steps Tags: Instagram, Facebook, Meta Graph API, Social Media, Token Refresh, Long-Lived Token, Marketing Automation, System User
By Brian Add data from a photo to Google Sheets
Automatically adding expense receipts to Google Sheets with Telegram, Mindee API, Twilio, and n8n.
By isa024787belCron routines with Telegram
Executes schedule routines, and triggers alerts via telegram
By HermilioBinance SM 1hour indicators tool
🧪 Binance SM 1hour Indicators Tool A precision trading signal engine that interprets 1-hour candlestick indicators for Binance Spot Market pairs using a GPT-4.1-mini LLM. Ideal for swing traders seeking directional bias and momentum clarity across medium timeframes. --- 🎥 Watch Tutorial: [](https://youtu.be/HWHD-eXP0IE) 🎯 Purpose This tool provides a structured 1-hour market read using: RSI (Relative Strength Index) MACD (Moving Average Convergence Divergence) BBANDS (Bollinger Bands) SMA & EMA (Simple and Exponential Moving Averages) ADX (Average Directional Index) It’s invoked as a sub-agent in broader AI workflows, such as the Binance Financial Analyst Tool and the Spot Market Quant AI Agent. --- ⚙️ Key Features | Feature | Description | | ---------------------- | ------------------------------------------------------------- | | 🔄 Subworkflow Trigger | Runs only when called by parent agent (not standalone) | | 🧠 GPT-4.1-mini LLM | Translates numeric indicators into natural-language summaries | | 📊 Real-time Data | Pulls latest 40×1h candles via internal webhook from Binance | | 📥 Input Format | { "message": "ETHUSDT", "sessionId": "telegramchatid" } | | 📤 Output Format | JSON summary + Telegram-friendly HTML overview | --- 💡 Example Output 📊 1h Technical Overview – ETHUSDT • RSI: 59 (Neutral) • MACD: Bullish Crossover • BBANDS: Price at Upper Band • EMA > SMA → Positive Slope • ADX: 28 → Moderate Trend Strength --- 🧩 Use Cases | Scenario | Result | | -------------------------------------- | ----------------------------------------------- | | Mid-frame market alignment | Verifies momentum between 15m and 4h timeframes | | Quant AI Agent input | Supplies trend context for entry/exit decisions | | Standalone medium-term signal snapshot | Validates swing trade setups or filters noise | --- 📦 Installation Instructions Import workflow into your n8n instance Confirm internal webhook /1h-indicators is live and authorized Insert your OpenAI credentials for GPT-4.1-mini node Use only when triggered via: Binance Financial Analyst Tool Binance Spot Market Quant AI Agent --- 🧾 Licensing & Support 🔗 Don Jayamaha – LinkedIn linkedin.com/in/donjayamahajr © 2025 Treasurium Capital Limited Company Architecture, prompts, and signal logic are proprietary. Redistribution or commercial use requires explicit licensing. No unauthorized cloning permitted.
By Don Jayamaha JrResearch-based article generator with OpenAI and web search for high SEO scores
Generate research-backed article with n8n Who’s it for Content marketers, SEO teams, and founders who need fast, research-grounded blog posts or long-form articles—multi-language included. Works well for teams that want citations, outlines, and section-by-section drafting with minimal manual effort. How it works / What it does Use a Form to collects domain, keywords, and target language. The workflow refines keywords, finds recent articles and authoritative citations, then synthesizes a master outline and loops through each section: generate search queries → fetch web results → summarize findings → write the section with the advanced model. Finally, it aggregates all sections into a clean Markdown article. Optional delivery nodes (Email, Telegram) and an AI Agent are included but disabled by default.&x20; How to set up Import the workflow JSON into n8n. Add your OpenAI credential Set simplemodel / advancedmodel in LLM Params Requirements n8n instance with outbound internet access. OpenAI API access (Responses API + websearchpreview). (Optional) Email/Telegram credentials if you want to deliver results async How to customize the workflow Edit prompts in LLM Params and Section Prompts to match tone, structure, and SEO style Tweak recency and source rules in Search Articles / Search Citations.&x20; Insert a human review step before “Write Section”, enable delivery nodes Change working/output languages in the Language node
By Constantine Kissel