Analyze Images & Extract Text with GPT-4o Vision and Telegram

This n8n workflow creates a fully functional Telegram vision bot that processes images sent by users. Upon receiving a photo message, it automatically fetches the highest-resolution version, converts it to base64 with normalized MIME type, and sends it to AIMLAPI's GPT-4o Vision endpoint using OpenAI-compatible messages format. The AI generates a short caption and extracts all visible text via OCR, which is then formatted and sent back to the original chat instantly.

Ideal for teams, makers, or customer support needing quick image analysis without building custom servers. It handles vision tasks like describing scenes, reading signs/documents, or identifying objects/text in photos. Benefits include zero-code setup, cost-effective AI usage (pay-per-use via AIMLAPI), and seamless Telegram integration for mobile-friendly interactions.

Use cases: Customer service bots for invoice scanning, content moderation for visual uploads, educational tools for diagram explanations, or personal assistants for photo summaries. Customize by tweaking prompts for specific domains (e.g., medical images, retail products), adding logic for text-only messages, logging interactions, or integrating rate limits/user whitelists.

Setup is straightforward: create a Telegram bot, add credentials to n8n, import the workflow, and test. Supports any vision-capable model on AIMLAPI, with options for retries/timeouts on complex images.