How to Upload HTML to NotebookLM
NotebookLM does not reliably handle HTML through its standard upload. Convert to Markdown first, then paste or upload the Markdown — that is the path with the fewest dead ends.
NotebookLM is happy to accept an HTML upload — it is the downstream behavior where things get interesting. NotebookLM does not accept HTML files natively in the standard upload flow, so you have to convert first. This guide walks through the whole path: where to drop the file, what usually breaks, and when it is worth converting to Markdown before you upload. Web pages bring their furniture with them. Navigation, cookie notices and footers all convert just as faithfully as the article you actually wanted. NotebookLM takes uploads via Add sources panel when creating or opening a notebook, caps individual files around 200MB, and works with roughly 500,000 tokens of context on its default variant. The short version: give NotebookLM clean structure — headings, lists, tables that survived the trip — rather than a raw HTML and the hope that the parser does the right thing.
Why convert to Markdown before uploading
The case for a Markdown intermediate step, specifically for HTML in NotebookLM.
- Markdown is token-efficient. You can fit substantially more of an HTML document inside NotebookLM's context window (currently around 500,000 tokens, give or take depending on the plan and model variant) once formatting overhead is gone.
- NotebookLM respects Markdown structure. Headings become navigable sections, lists stay as lists, and tables that survive conversion stay readable for the model.
- You get a predictable failure mode. When an HTML parser inside NotebookLM silently drops a page or column, you rarely notice until the answer is wrong. With Markdown, what you see is what the model sees.
- Batch and reuse get easier. A Markdown file drops cleanly into Custom GPTs, Claude Projects, NotebookLM notebooks, or a RAG pipeline without per-tool quirks.
- HTML uploads carry nav bars, cookie banners, and inline scripts that pollute the context. Markdown conversion strips the chrome and keeps the actual content.
Markdown is one of the few formats NotebookLM accepts natively, and the clean structure gives you better citations and source previews. Converted HTML files come through at around 98% text accuracy, with table fidelity we would call excellent. The usual catch with HTML: javaScript-rendered content requires pre-processing. Convert once and reuse the Markdown across NotebookLM chats rather than re-uploading the original every time.
Step-by-step: HTML into NotebookLM
- 1
Skip the direct HTML upload
NotebookLM does not reliably accept HTML in the standard flow. Do not waste time wrestling with it — go straight to a Markdown conversion.
- 2
Convert the HTML file to Markdown
Open markdownconverters.com/conversion/html, drop your HTML file in, and download the .md output. Strip navigation and cookie banners first if you can. Otherwise the boilerplate ends up in your Markdown and competes with the actual content.
- 3
Skim the Markdown before you upload
Open the output in any text editor. Check that the main article survived and that navigation lists did not end up as content. Thirty seconds here saves an hour of debugging answers that were never going to be right.
- 4
Upload to NotebookLM
In notebooklm.google.com, open or create a notebook, click "Add sources", and upload the Markdown file. NotebookLM natively supports .md, so no further conversion is needed.
- 5
Prompt with structure in mind
Reference the file by name, then point at a section rather than asking broadly. That means quoting the heading: "In the 'Pricing' section, what is the tier for..." instead of "what does this say about pricing". NotebookLM retrieves noticeably better when you narrow it yourself.
- 6
Verify against the source
Ask NotebookLM to quote the passage it is citing. If the quote does not appear in the Markdown, you have either a retrieval miss or a hallucination — re-check the conversion or narrow the prompt.
Gotchas with HTML in NotebookLM
The specific failure modes we see most often — worth knowing before you spend an afternoon debugging "bad" model answers that are really upload problems.
- NotebookLM wants URLs, not HTML files
NotebookLM accepts website URLs as sources, but it does not take an .html upload. If you need an offline page in a notebook, convert the HTML to Markdown first.
- NotebookLM does not accept DOCX, XLSX, or PPTX directly
you have to convert to PDF, Markdown, or TXT first
- Each notebook has a source cap (currently dozens on free, more on Plus)
verify the active limit before bulk-adding sources
- Tables embedded in PDFs are flattened into prose; numeric analysis becomes unreliable
Tables embedded in PDFs are flattened into prose; numeric analysis becomes unreliable
- HTML uploads are not supported; you must paste a URL or convert the page
HTML uploads are not supported; you must paste a URL or convert the page
Getting better answers from NotebookLM
- Upload through Add sources panel when creating or opening a notebook — that is the path Google actually supports, and the one least likely to silently truncate your file.
- NotebookLM holds roughly 500,000 tokens, about 364k words. Converted HTML content is compact, but a long document plus a long conversation still competes for the same budget.
- Stay under about 10MB per file even though the hard cap is 200MB. Retrieval quality degrades well before the limit, particularly on dense HTML content.
- HTML uploads are not supported; you must paste a URL or convert the page.
- Each notebook has a source cap (currently dozens on free, more on Plus) — verify the active limit before bulk-adding sources.
- NotebookLM reads Markdown structure well, so ask it to cite the heading it answered from. If it cannot name one, your content was not retrieved.
- Convert the article URL rather than a saved page where you can. Saved HTML tends to carry inlined scripts and styling that survive as noise in the Markdown.
Frequently asked questions
Can I upload an HTML file to NotebookLM directly?
Not reliably. NotebookLM either does not accept HTML at all or accepts it with heavy restrictions. The supported path is to convert to Markdown (or PDF, in a pinch) and upload that.
What is the file-size limit on NotebookLM?
As of this writing, NotebookLM typically caps individual uploads around 200MB, with a context window of roughly 500,000 tokens on the default variant. These numbers shift with releases and differ by plan, so verify in the app before relying on a specific figure. The practical limit is lower than the hard cap anyway — retrieval quality starts to degrade well before you hit it, especially for dense technical documents. Smaller, focused files beat one enormous one.
Does converting HTML to Markdown lose information?
Markdown drops visual formatting — fonts, colors, exact page layout — because it is a semantic format. It preserves headings, lists, tables, and text content, which is what LLMs actually use. For HTML specifically, the things that matter (Web scraping projects, Blog content migration) survive cleanly. Things like decorative styling do not, and usually should not.
How many HTML files can I upload to NotebookLM at once?
NotebookLM caps each notebook at a fixed number of sources (the exact count depends on tier — free vs. Plus). Converting a batch of HTML files to Markdown first lets you pack more content under that source count by consolidating.
Is this safe — what happens to my HTML file?
When you upload to NotebookLM, the content is processed by Google's infrastructure under their terms. If the document is sensitive, consider using the enterprise tier (where retention and training policies differ), or convert locally where possible. Markdown Converters does not retain your uploaded files beyond the conversion window.
Keep going
Drop HTML files in and download Markdown ready to upload to NotebookLM.
Check how much of your converted HTML will actually fit inside NotebookLM's context window.
Same workflow, tuned for ChatGPT's upload quirks and context window.
Same workflow, tuned for Claude's upload quirks and context window.
Working with PDF instead of HTML? The gotchas are different.
Background on why clean Markdown consistently outperforms native uploads across every major assistant.
Convert HTML for NotebookLM in one step
Clean Markdown output ready to drop into NotebookLM. Free tier, no account required to try.