Supported File Types

Artifact supports the following file types for parsing documents and building catalogs (RAG):

• .pdf - Portable Document Format
• .doc - Microsoft Word 97-2003
• .docx - Microsoft Word 2007-2019
• .txt - Plain text
• .md - Markdown
• .html - HTML document
• .ppt - Microsoft PowerPoint 97-2003
• .pptx - Microsoft PowerPoint 2007-2019
• .xls - Microsoft Excel 97-2003
• .xlsx - Microsoft Excel 2007-2019
• .csv - Comma-separated values

📘

Max file size: 512 MB per file

Parsing Behavior

• All supported formats are converted into clean, structured Markdown for use with LLMs — except .txt files, which are extracted as plain text.
• Scanned and handwritten PDFs are supported with high-quality visual understanding.