Turn any PDF into a clean Google Doc with Mistral OCR

by Hunyao ✓ · 2,591 views · created 2025-08-03 · Document Extraction, Multimodal AI

What this workflow does

This AI workflow has 15 nodes and runs when someone submits an n8n form (“Document Scanner”). Along the way it connects Google Docs and OpenRouter Chat Model. It also calls api.mistral.ai over HTTP.

How it runs

Trigger: When someone submits an n8n form (“Document Scanner”).

1. Upload • PDF to Mistral (HTTP Request): calls api.mistral.ai (POST)
2. Fetch • Signed URL (HTTP Request): calls an HTTP API
3. OCR • PDF via Mistral (HTTP Request): calls api.mistral.ai (POST)
4. Extract • Image Placeholders (Code): runs custom JavaScript code
5. Merge: merges the data from its input branches
6. OCR • Inline Images (HTTP Request): calls api.mistral.ai (POST)
7. Merge: merges the data from its input branches
8. Enrich • Replace Placeholders (Code): runs custom JavaScript code
9. Set • Rename Columns (Edit Fields (Set)): sets 4 fields: pageIndex, imageId, imageDataUri…
10. Clean • Markdown Text (Basic LLM Chain): sends a prompt to an LLM, with model: LLM Model • GPT-4.1-mini

…plus 3 more steps. See the full node list below.

Services it uses

Google DocsOpenRouter Chat Model

HTTP calls to: api.mistral.ai

Before you upgrade to n8n 3.0

Nodes used (9)

• n8n Form Triggern8n-nodes-base.formTrigger
• HTTP Requestn8n-nodes-base.httpRequest
• Coden8n-nodes-base.code
• Mergen8n-nodes-base.merge
• Edit Fields (Set)n8n-nodes-base.set
• Aggregaten8n-nodes-base.aggregate
• Basic LLM Chain@n8n/n8n-nodes-langchain.chainLlm
• Google Docsn8n-nodes-base.googleDocs
• OpenRouter Chat Model@n8n/n8n-nodes-langchain.lmChatOpenRouter

From the author

Upload a PDF and instantly get a neatly formatted Google Doc with all the readable text—no manual copy-paste, no messy line breaks. What this workflow does Accepts PDF uploads via a public form Sends the file to Mistral Cloud for high-accuracy OCR Detects and…

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