Turn any image link into clean JSON you can use right away. It helps teams capture details from IDs, invoices, and receipts without manual typing. Great for operations that need fast and accurate data entry.

A public webhook receives a request with an image URL and a list of fields to extract. The flow downloads the image, converts it to base64, and sends it to the Gemini API Flash Lite model for OCR and understanding. The model returns text that is shaped into a simple JSON object with only the fields you asked for. The response is sent back to the caller in real time, so it can plug into forms, back office tools, or custom apps.

You only need a working n8n instance and a Gemini API key. Expect to cut data entry time from minutes to seconds for each document and reduce copy paste mistakes. Common uses include ID verification, invoice data capture, receipt logging, and business card parsing for quick intake in your system.

• Public webhook endpoint that accepts an image URL and your required fields
• Automatic image download from the provided URL
• Image converted to base64 so the AI model can process it
• Gemini API call with inline image data for OCR and data extraction
• Custom field mapping using a properties object to control the output schema
• Response cleaned to a single JSON object and sent back instantly

• Reduce manual data entry from 10 minutes to under 1 minute per document
• Improve accuracy by up to 90 percent by removing copy paste steps
• Handle 10 times more requests with a simple API endpoint
• Return only the fields you need to keep downstream data clean
• Connect any app that can call a webhook for easy rollout

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with Gemini API. See the Tools Required section above for links to create accounts with these services.
3. Open the Webhook node and note the path value. Switch to the Production URL when you are ready and set the response mode to use the Respond to Webhook node.
4. Create your Gemini API key in your Google AI account. In n8n, double click the Call Gemini API node, then on the Credential to connect with dropdown, click Create new credential, choose Google PaLM API, and paste your API key. Save the credential.
5. In the Get image from URL node, confirm the URL field uses the expression {{$json.body.image_url}} so the image link from the request is downloaded.
6. Check the Transform image to base64 node and keep the destination key as data1 so the encoded image is available to the AI request.
7. Open the Call Gemini API node and verify the endpoint and model settings. If you send PNG images, change the mimeType to image/png in the inlineData section.
8. In the Set node, keep the result expression so it parses the model text into JSON and returns only the fields you requested.
9. Activate the workflow. Send a test GET request to the webhook URL with image_url, Requirement, and properties in the body. Use the sample from the notes as a guide.
10. If you get a 401 error, recheck the Gemini API key in the credential. If the response is empty, make sure the image URL is public and the mime type matches the image format.