Extract and process information directly from PDF using Claude and Gemini
Workflow JSON
{"meta": {"instanceId": "f4f5d195bb2162a0972f737368404b18be694648d365d6c6771d7b4909d28167"}, "nodes": [{"id": "b6cd232e-e82e-457b-9f03-c010b3eba148", "name": "When clicking 'Test workflow'", "type": "n8n-nodes-base.manualTrigger", "position": [-40, 0], "parameters": {}, "typeVersion": 1}, {"id": "2b734806-e3c0-4552-a491-54ca846ed3ac", "name": "Extract from File", "type": "n8n-nodes-base.extractFromFile", "position": [620, 0], "parameters": {"options": {}, "operation": "binaryToPropery"}, "typeVersion": 1}, {"id": "2c199499-cc4f-405c-8560-765500b7acba", "name": "Google Drive", "type": "n8n-nodes-base.googleDrive", "position": [420, 0], "parameters": {"fileId": {"__rl": true, "mode": "list", "value": "18Ac2xorxirIBm9FNFDDB5aVUSPBCCg1U", "cachedResultUrl": "https://drive.google.com/file/d/18Ac2xorxirIBm9FNFDDB5aVUSPBCCg1U/view?usp=drivesdk", "cachedResultName": "Invoice-798FE2FA-0004.pdf"}, "options": {}, "operation": "download"}, "credentials": {"googleDriveOAuth2Api": {"id": "", "name": "[Your googleDriveOAuth2Api]"}}, "typeVersion": 3}, {"id": "e3031c0c-f059-4f30-9684-10014a277d55", "name": "Call Gemini 2.0 Flash with PDF Capabilities", "type": "n8n-nodes-base.httpRequest", "position": [880, 220], "parameters": {"url": "https://generativelanguage.googleapis.com/v1beta/models/gemini-2.0-flash-exp:generateContent", "method": "POST", "options": {}, "jsonBody": "={\n \"contents\": [\n {\n \"parts\": [\n {\n \"inline_data\": {\n \"mime_type\": \"application/pdf\",\n \"data\": \"{{ $json.data }}\"\n }\n },\n {\n \"text\": \"{{ $('Define Prompt').item.json.prompt }}\"\n }\n ]\n }\n ]\n}", "sendBody": true, "specifyBody": "json", "authentication": "predefinedCredentialType", "nodeCredentialType": "googlePalmApi"}, "credentials": {"anthropicApi": {"id": "", "name": "[Your anthropicApi]"}, "googlePalmApi": {"id": "", "name": "[Your googlePalmApi]"}}, "typeVersion": 4.2}, {"id": "135df716-32a1-47e8-9ed8-30c830b803d6", "name": "Call Claude 3.5 Sonnet with PDF Capabilities", "type": "n8n-nodes-base.httpRequest", "position": [880, -140], "parameters": {"url": "https://api.anthropic.com/v1/messages", "method": "POST", "options": {}, "jsonBody": "={\n \"model\": \"claude-3-5-sonnet-20241022\",\n \"max_tokens\": 1024,\n \"messages\": [{\n \"role\": \"user\",\n \"content\": [{\n \"type\": \"document\",\n \"source\": {\n \"type\": \"base64\",\n \"media_type\": \"application/pdf\",\n \"data\": \"{{$json.data}}\"\n }\n },\n {\n \"type\": \"text\",\n \"text\": \"{{ $('Define Prompt').item.json.prompt }}\"\n }]\n }]\n}", "sendBody": true, "sendHeaders": true, "specifyBody": "json", "authentication": "predefinedCredentialType", "headerParameters": {"parameters": [{"name": "anthropic-version", "value": "2023-06-01"}, {"name": "content-type", "value": "application/json"}]}, "nodeCredentialType": "anthropicApi"}, "credentials": {"anthropicApi": {"id": "", "name": "[Your anthropicApi]"}}, "typeVersion": 4.2}, {"id": "5b8994d1-4bfd-4776-84ac-b3141aca6378", "name": "Sticky Note1", "type": "n8n-nodes-base.stickyNote", "position": [-700, -280], "parameters": {"color": 7, "width": 601, "height": 585, "content": "## Workflow: Extract data from PDF with Claude 3.5 Sonnet or Gemini 2.0 Flash\n\n**Overview**\n- This workflow helps you compare Claude 3.5 Sonnet and Gemini 2.0 Flash when extracting data from a PDF\n- This workflow extracts and processes the data within a PDF in **one single step**, **instead of calling an OCR and then an LLM\u201d**\n\n\n**How it works**\n- The initial 2 steps download the PDF and convert it to base64.\n- This base64 string is then sent to both Claude 3.5 Sonnet and Gemini 2.0 Flash to extract information.\n- This workflow is made to let you compare results, latency, and cost (in their dedicated dashboard).\n\n\n**How to use it**\n- Set up your Google Drive if not already done\n- Select a document on your Google Drive\n- Modify the prompt in \"Define Prompt\" to extract the information you need and transform it as wanted.\n- Get a [Claude API key](https://console.anthropic.com/settings/keys) and/or [Gemini API key](https://aistudio.google.com/app/apikey)\n- Note that you can deactivate one of the 2 API calls if you don't want to try both\n- Test the Workflow\n"}, "typeVersion": 1}, {"id": "616241a9-6199-406b-88dc-0afc7d974250", "name": "Sticky Note", "type": "n8n-nodes-base.stickyNote", "position": [820, 60], "parameters": {"color": 5, "width": 320, "height": 360, "content": "You can output the result as JSON by adding the following:\n```\n\"generationConfig\": {\n \"responseMimeType\": \"application/json\"\n```\nor even use a structured output.\n[Check the documentation](https://ai.google.dev/gemini-api/docs/structured-output?lang=rest)"}, "typeVersion": 1}, {"id": "bbac8d3d-d68f-4aa2-a41a-b06f7de2317b", "name": "Define Prompt", "type": "n8n-nodes-base.set", "position": [180, 0], "parameters": {"options": {}, "assignments": {"assignments": [{"id": "dba23ef5-95df-496a-8e24-c7c1544533d2", "name": "prompt", "type": "string", "value": "Extract the VAT numbers for each country"}]}}, "typeVersion": 3.4}, {"id": "3c2e7265-76e5-4911-a950-7e6b0c89ec5a", "name": "Sticky Note2", "type": "n8n-nodes-base.stickyNote", "position": [820, -200], "parameters": {"color": 5, "width": 320, "height": 240, "content": "You can force Claude to output JSON with [Prefill response format](https://docs.anthropic.com/en/docs/test-and-evaluate/strengthen-guardrails/increase-consistency#prefill-claudes-response)"}, "typeVersion": 1}, {"id": "f2b46305-5200-486e-ad4d-ecc0d2a14314", "name": "Sticky Note3", "type": "n8n-nodes-base.stickyNote", "position": [380, -120], "parameters": {"color": 5, "width": 380, "height": 280, "content": "These 2 steps first download the PDF file, and then convert it to base64.\nThis is required by both APIs to process the file."}, "typeVersion": 1}, {"id": "e5dff70f-b55a-4c23-9025-765a7cf19c4a", "name": "Sticky Note4", "type": "n8n-nodes-base.stickyNote", "position": [120, -120], "parameters": {"color": 5, "width": 220, "height": 280, "content": "This prompt is used in both Gemini\u2019s and Claude\u2019s calls to define what information should be extracted and processed."}, "typeVersion": 1}], "pinData": {}, "connections": {"Google Drive": {"main": [[{"node": "Extract from File", "type": "main", "index": 0}]]}, "Define Prompt": {"main": [[{"node": "Google Drive", "type": "main", "index": 0}]]}, "Extract from File": {"main": [[{"node": "Call Claude 3.5 Sonnet with PDF Capabilities", "type": "main", "index": 0}, {"node": "Call Gemini 2.0 Flash with PDF Capabilities", "type": "main", "index": 0}]]}, "When clicking 'Test workflow'": {"main": [[{"node": "Define Prompt", "type": "main", "index": 0}]]}}}How to Import This Workflow
- 1Copy the workflow JSON above using the Copy Workflow JSON button.
- 2Open your n8n instance and go to Workflows.
- 3Click Import from JSON and paste the copied workflow.
Don't have an n8n instance? Start your free trial at n8nautomation.cloud
Related Templates
ETL pipeline
Automate your data extraction, transformation, and loading with this robust ETL pipeline, designed to efficiently process and analyze information from various sources. This workflow begins on a schedule, fetching tweets from Twitter/X, then storing them in MongoDB for initial processing. The MongoDB data is then sent to Google Cloud Natural Language for sentiment analysis or entity extraction, with the results subsequently prepared and stored in PostgreSQL. A conditional check on the PostgreSQL data determines whether to send an alert to Slack, ensuring timely notifications for critical insights or anomalies. This powerful automation is ideal for marketing teams monitoring brand sentiment, researchers analyzing public opinion, or businesses tracking competitor activity, providing actionable intelligence without manual data handling. By automating data ingestion, enrichment, and storage, this workflow significantly reduces the time and effort spent on data preparation, allowing teams to focus on analysis and strategic decision-making while ensuring data consistency and accessibility.
SQL agent with memory
Empower your data analysis with the SQL agent with memory workflow, automating the process of querying databases using natural language. This powerful workflow connects OpenAI's advanced language models with your local SQL databases, allowing you to interact with your data through a conversational interface. Initially, the workflow downloads a chinook.zip example database, extracts it, and saves the chinook.db file locally, making it immediately available for querying. The AI Agent, powered by OpenAI Chat Model and supported by a Window Buffer Memory, interprets your natural language questions, translates them into SQL queries, executes them against your local chinook.db, and provides the results back to you. This is incredibly useful for data analysts, business intelligence professionals, or anyone needing quick insights from their databases without writing complex SQL queries, significantly reducing the time and specialized knowledge required for data exploration. By leveraging the Chat Trigger, users can easily initiate conversations and receive immediate, intelligent responses, streamlining data access and accelerating decision-making.
Prepare CSV files with GPT-4
Transform raw, unstructured text into perfectly formatted CSV files using the power of GPT-4 with this n8n workflow. This automation connects OpenAI's advanced language model to process your input, then meticulously structures the output into a usable CSV format. Ideal for data analysts, marketers, or researchers, this workflow helps you extract specific information from large text datasets, such as customer reviews, survey responses, or article summaries, and prepare it for analysis in spreadsheets or databases. By automating the extraction and formatting of data, you significantly reduce manual data entry errors and save countless hours of tedious work, allowing you to focus on insights rather than data preparation. The workflow manually triggers, sending your text to OpenAI, then splits the responses into manageable batches, parses the JSON output, converts it into a structured table, and finally saves a clean, UTF-8 encoded CSV file to disk, ensuring compatibility across various systems.