Getting Started

1. Create an account

Head to app.duckviz.com and sign in with Google or email. New accounts receive free credits automatically — no payment required to try the product.

2. Upload a file

From the Upload screen, choose one of:

• File — drop a CSV, Excel, JSON, XML, or log file (up to several hundred MB).
• Folder — drop an entire directory; DuckViz will walk it, skip binaries and dotfiles, and ingest each supported file into its own DuckDB table.
• Fetch from API — provide a URL (with optional headers and body) and DuckViz fetches the response through its proxy.

Your file is parsed locally in the browser by DuckDB-WASM. Nothing is uploaded to our servers.

3. Let the AI build widgets

As soon as the table is ready, DuckViz's widget pipeline runs:

1. Classifies the data domain (e.g. "Sales", "Network & Firewall", "Web traffic logs").
2. Picks 5 chart types from a 70+ registry based on your actual column shapes.
3. Generates a DuckDB SQL query per chart and renders it with a pre-built D3 component.

You'll see charts stream in one by one. Drag any chart onto your dashboard.

4. Generate a report

Open the Dashboard and click Generate Report. DuckViz picks a template automatically based on your data domain, streams an executive summary in a Notion-style editor, and renders live charts inside the document. Export as PDF, DOCX, or PPTX when you're done.

Next steps

• Read Privacy-first architecture to understand what leaves the browser and what doesn't.
• Install the CLI to push files from your terminal instead of clicking around.