Upload the PDF
Choose the source document. Pages without extractable text are surfaced as warnings.
Gixo Lumen · deterministic document conversion
The converter extracts PDF text and available structure into editable PowerPoint slides. Download the PPTX after sign-in, then import it into Google Slides yourself; no Google API or account connection is involved.
A converter, not a prompt box
The useful boundary is simple: make an editable PPTX, then let Google Slides import it through its own supported workflow.
The conversion step is deliberately structural. It extracts the source, identifies headings and blocks, and splits long sections into readable slides — it does not paraphrase your source or invent a narrative around it. Rewriting, restructuring and theming are separate, opt-in steps you take afterwards on the canvas.
Free preview: convert and inspect the complete deck in your browser without an account. Sign in is the download boundary for PPTX and PDF.
How it works
Choose the source document. Pages without extractable text are surfaced as warnings.
Headings, blocks, tables, and available images are assembled into a real PPTX.
Review the deck, sign in to download the PPTX, then use Google Slides' normal import flow.
Source rules
PDF extraction and slide assembly are deterministic. The output is an ordinary editable PPTX intended for manual import into Google Slides.
Google Slides is not called. The output is a normal PPTX you can import yourself.
PDF pages and package output are validated and bounded before the preview is built.
Scanned or image-only pages are detected and reported rather than silently OCR'd.
Output behavior
Supported formats & specifications
PDF upload cap: 6 MiB. The 57.95 KB text PDF fixture produced a 1-slide, 4.98 KB editable PPTX; scanned-page OCR and ~10 MB timing figures are outside this batch.
| Aspect | Current behavior |
|---|---|
| Input | |
| Result | A full editable preview and an import-ready PPTX report with source-page and warning details. |
| Free vs signed in | The preview and report are free without an account. Sign in to download the PPTX or PDF; importing into Google Slides remains your manual step. |
| Measured limits | PDF upload cap: 6 MiB. The 57.95 KB text PDF fixture produced a 1-slide, 4.98 KB editable PPTX; scanned-page OCR and ~10 MB timing figures are outside this batch. |
| Extraction | Structural, not generative: headings, blocks and tables are read from the source rather than rewritten. |
| Safety | Uploads are validated and bounded-parsed. URL conversion accepts public HTTP/HTTPS pages only. |
What conversion does, and what it leaves to you
Conversion reads the structure already in your source and lays it out as slides. The preview is free without an account; sign in for the PPTX or PDF download, then import the ordinary PPTX into Google Slides yourself. Nothing is silently reworded between the file you brought and the deck you present.
A note about Google Slides: This tool does not log in to Google or upload to Google Slides. It gives you a normal PPTX and keeps that handoff explicit.
Current public contract
| Aspect | Current behavior | Why it matters |
|---|---|---|
| Input | The named source front-end determines which structural signals are read. | |
| Review | Every slide is editable on the canvas before export. | You can inspect and correct the output before anything leaves the workspace. |
| Free preview | Full editable deck preview without an account. | Slide count, source-page count, and extraction warnings are visible before sign-in. |
| Export | PPTX and PDF after sign-in, with editable text and native tables. | Slides are rebuilt as real objects rather than flattened page images. |
| Method | Structural extraction and slide assembly, not paraphrase. | The same source structure produces the same conversion behavior. |
No. It produces an ordinary editable PPTX that you import into Google Slides yourself.
It warns when pages contain no extractable text. OCR is outside this deterministic batch.
The output is built as a real PPTX with native text and tables where extraction exposes them; review it before import.
No. There is no Google connection and no LLM call in the conversion path.
Questions people ask before converting
No. It produces an ordinary editable PPTX that you import into Google Slides yourself.
It warns when pages contain no extractable text. OCR is outside this deterministic batch.
The output is built as a real PPTX with native text and tables where extraction exposes them; review it before import.
No. There is no Google connection and no LLM call in the conversion path.
Ready to inspect the result?
Reviewing is the point: you can see whether the source structure made a useful deck, and fix it on the canvas, before anything is exported.
Convert PDF with Gixo Lumen