Jupyter → SageMath HTML Converter

1Upload the SageMath HTML template

A blank notebook with your custom settings (e.g., theme, your own Sage Cell server, AI settings etc.). These settings will be applied to all the exported notebooks. Current version of a blank SageMath AI notebook (v20) can be found at GitHub .

📄

Drop your template .html here

or click to browse

2Add notebooks, a ZIP archive or a folder

Three ways to feed the converter — they can be combined: (a) one or more loose .ipynb files, (b) a .zip archive with any folder structure, (c) a folder dragged in or picked with the folder button. Inside archives/folders only .ipynb files are converted; everything else (images, data, PDFs…) is copied over unchanged.

📓 📦 📁

Drop .ipynb files, a .zip, or a folder here

or click to browse files (multiple allowed)

3Options, convert & download

Output naming Use the notebook file name (default) Lecture_03.ipynb → Lecture_03.html. Recommended when the exported files must match your notebook names. Use the first H1 heading The name produced by the template's own saveHtml(), e.g. # Linear algebra → Linear algebra.html. Duplicates get _1, _2… suffixes.

Archive / folder mode Keep the original .ipynb files in the output Off by default — normally only the generated HTML files plus all non-notebook files are exported. Ignore .ipynb_checkpoints (and __MACOSX, .DS_Store) Auto-save copies are neither converted nor copied to the output.

🗑️ Stale-HTML rule (always on): if a folder already contains an HTML file whose name matches a notebook in that same folder (notes.ipynb ↔ notes.html), that old HTML is replaced by the fresh export. All other HTML files in the folder are left untouched.

Settle time (ms) Per-file delay (ms) Zip filename

How it works

The converter uses your template — a blank SageMath HTML notebook — as the host environment. The template is loaded into a hidden iframe, which means all the custom settings are applied in the exported HTML notebooks. Nothing is uploaded anywhere: reading, converting and zipping happen locally in your browser.

What you get back

One loose notebook → the .html file is downloaded directly.
Several loose notebooks → one .zip containing the HTML files.
ZIP / folder input → one .zip with the original folder structure, the generated HTML files, and all non-notebook files preserved.

Example (archive mode, defaults)

input.zip                       output.zip
├─ course/                      ├─ course/
│   ├─ intro.ipynb              │   ├─ intro.html      ← new
│   ├─ intro.html   (old)       │   ├─ syllabus.html   ← untouched
│   ├─ syllabus.html            │   └─ img/plot.png    ← copied
│   └─ img/plot.png             └─ data/values.csv     ← copied
└─ data/values.csv

intro.html was regenerated (the stale one was dropped), syllabus.html had no matching notebook so it stayed, and intro.ipynb was omitted because “Keep the original .ipynb files” is off.

Naming: file name vs. H1 heading

File name (default) — the export is named exactly like the notebook, so the set of file names in your folders stays stable. Best for archives/folders, since links and references usually rely on the notebook name.
H1 heading — the legacy behaviour: the title from the notebook's first # heading is used. In archive mode the file is still written into the notebook's own folder; if the H1 name collides with a file produced in the same run, a numeric suffix is added.

Notes & limits

Folder drag & drop works in Chromium-based browsers and Firefox; if a dropped folder isn't picked up, use the “Select a folder instead…” button.
Multiple archives/folders at once: each archive's contents are placed inside a top-level folder named after the archive, so their trees can't clash. A single archive keeps its structure at the root of the output zip.
If a notebook fails to convert, its original .ipynb is kept in the output and no existing HTML is deleted — you never lose data because of a failed run.
Encrypted or unusually compressed ZIPs cannot be read in the browser; unzip them first.
Everything is processed sequentially, so very large archives simply take longer.

Timing settings

Two timers control how long the converter waits between operations inside the iframe. They exist because rendering happens asynchronously since the page (as of now) can't know in advance when it's truly done. Defaults work for typical notebooks — increase them if you see incomplete output, decrease them to speed up batches of small files.

Setting	What it controls	When to change it
Settle time	How long to wait after `.ipynb` import before triggering the Save sequence (as if one would do it in the Sage notebook).	Increase (e.g. 4000–6000 ms) for long notebooks with lots of LaTeX or large outputs. Decrease (e.g. 1500 ms) for short notebooks if you want batches to finish faster.
Per-file delay	Pause between consecutive conversions, after one is finished and before the next iframe is created.	Usually fine at the default. Increase if your browser seems sluggish during big batches; set to 0 for maximum speed.