💬 Process telegram image albums with Data Tables Cache and NanoBanana AI

⚡ 895 views · 💬 Support Chatbots

Description

This n8n workflow template simplifies processing of media group messages sent by users in Telegram. It caches image album messages using Data Tables, and then sends them to a Nano Banana model for processing.

This template showcases n8n’s new Data Tables feature as a powerful caching layer to process entire image albums from Telegram.

How it works

• Media Group handling: captures all images and the shared caption when user sends an album via Telegram.
• Data Table caching: stores all incoming messages in a Data Table to prevent data loss and allow for batch processing.
• Scheduled processing: a timer trigger periodically finds and processes completed image albums from the cache.
• Multi-modal AI: sends the full set of images and the text prompt to the NanoBanana model via OpenRouter.
• User feedback: automatically notifies the user when processing begins and when the result is ready.

How to use

• Import the workflow template into your n8n instance.
• Create a new Data Table with the following columns (all of type String): chat_id, message_id, media_group, message, status.
• Configure your Telegram credentials and connect your bot.
• Update the “prepare user messages” node with your Telegram bot token to create the image download links.
• Set up your OpenRouter credentials.
• Activate the workflow. Now you can send a group of images with a caption to your Telegram bot.

Requirements

• An n8n instance (cloud or self-hosted) with Data Tables enabled. Minimal n8n version is 1.113
• A Telegram account and a Telegram bot token.
• An OpenRouter account with access to the NanoBanana model.

Customizing this workflow

• Adjust the processing schedule: modify the Schedule Trigger node to change how often the workflow checks for new media groups to process.
• Handle other message types: The workflow handles only the media groups. You can easily build logic for text messages, documents, or commands by adding nodes instead of the placeholder node (“Process other messages as usual”).
• Extend the functionality: add more nodes after the AI generates an image to perform different actions, such as saving the file, sending it to another service, or creating a social media post.

🔗 Nodes Used

HTTP Request, Telegram, Telegram Trigger, Schedule Trigger, Filter, Convert to File

📥 Import

Download workflow.json and import into n8n: Workflow menu → Import from File

📖 Importing guide · 🔑 Credential setup