Create a multi-modal Telegram support bot with GPT-4 and Supabase RAG

by Ezema Kingsley Chibuzo ✓ · 11,909 views · created 2025-07-02 · Support Chatbot, AI RAG

What this workflow does

This AI workflow has 48 nodes and runs manually, when you click “Execute workflow” (plus 1 other trigger). Along the way it connects Telegram, OpenAI, Google Drive and OpenAI Chat Model and 4 more. It also calls v2.convertapi.com over HTTP.

How it runs

Trigger: Manually, when you click “Execute workflow” (plus 1 other trigger).

1. Download file (Google Drive): download
2. Input Message Router (Switch): routes items into 4 branches
3. Typing… (Telegram): send chat action
4. Add to Supabase Vector DB (Supabase Vector Store): stores documents in the Supabase vector store, with embeddings: Embeddings OpenAI; document loader: Default Data Loader
5. Extract Text (Edit Fields (Set)): sets 1 field: text
6. Download Audio (Telegram): work with file
7. Download Photo (Telegram): work with file
8. Supported Document File Types (Code): runs custom JavaScript code
9. Knowledge Base AI Agent (AI Agent): runs an AI agent, with model: OpenAI Chat Model; 1 tool (Supabase Vector Store); memory: Postgres Chat Memory
10. Translate to Text (OpenAI): translate (audio)

…plus 28 more steps. See the full node list below.

Services it uses

TelegramOpenAIGoogle DriveOpenAI Chat ModelEmbeddings OpenAIPostgres Chat MemoryReranker CohereSupabase Vector Store

HTTP calls to: v2.convertapi.com

Before you upgrade to n8n 3.0

Nodes used (19)

• Telegram Triggern8n-nodes-base.telegramTrigger
• Telegramn8n-nodes-base.telegram
• OpenAI@n8n/n8n-nodes-langchain.openAi
• Coden8n-nodes-base.code
• Ifn8n-nodes-base.if
• Extract from Filen8n-nodes-base.extractFromFile
• Switchn8n-nodes-base.switch
• HTTP Requestn8n-nodes-base.httpRequest
• Manual Triggern8n-nodes-base.manualTrigger
• Google Driven8n-nodes-base.googleDrive
• Default Data Loader@n8n/n8n-nodes-langchain.documentDefaultDataLoader
• Recursive Character Text Splitter@n8n/n8n-nodes-langchain.textSplitterRecursiveCharacterTextSplitter
• OpenAI Chat Model@n8n/n8n-nodes-langchain.lmChatOpenAi
• Embeddings OpenAI@n8n/n8n-nodes-langchain.embeddingsOpenAi
• Postgres Chat Memory@n8n/n8n-nodes-langchain.memoryPostgresChat
• Reranker Cohere@n8n/n8n-nodes-langchain.rerankerCohere
• Edit Fields (Set)n8n-nodes-base.set
• Supabase Vector Store@n8n/n8n-nodes-langchain.vectorStoreSupabase
• AI Agent@n8n/n8n-nodes-langchain.agent

From the author

🧠 What It Does This n8n workflow turns your Telegram bot into a smart, multi-modal AI assistant that accepts text, documents, images, and audio messages, interprets them using OpenAI models, and responds instantly with context-aware answers. It integrates a…

View the original on n8n.io → Check your own copy