Supported File Types
Smartchat extracts the content of a file with a parser built for its format, then indexes it with an embedding model so it can be found during chats. Only formats with a parser can be ingested — this page lists them.
The same list applies everywhere a file enters a Space: manual upload, S3 AutoSync, and Confluence Sync.
Formats
Section titled “Formats”| Category | File types |
|---|---|
| Text documents | pdf, doc, docx, rtf, epub |
| Presentations | ppt, pptx |
| Spreadsheets | xls, xlsx, csv |
| Text and markup | txt, md, html, xml, json |
msg, mbox | |
| Source code | py, pyw, pyi, js, mjs, cjs, jsx, ts, tsx, mts, go, rs, java, c, h, cpp, hpp, cc, cs, rb, php, swift, kt, scala, sh, bash, sql |
| Audio | mp3 |
Audio files are transcribed to text before indexing.
If a format is not listed
Section titled “If a format is not listed”Convert the file to a supported format before uploading it — for example, export to PDF or DOCX. If ingestion fails for a file whose format is listed, see Ingestion Errors.
See also
Section titled “See also”- File Size Limits — file size, page and text volume cutoffs