NotebookLM Review (2026): Now Called Gemini Notebook

Google renamed NotebookLM to Gemini Notebook on July 16, 2026, folding it fully into the Gemini brand — new logo, deeper Gemini 3 integration, same core product. If you upload your own documents and want an AI that only answers from those sources (with citations back to the exact passage), this is still the best tool built specifically for that job.

Updated September 1, 2026: this page reflects the current model lineup, including GPT-5.6 (flagship Sol, mid-tier Terra, budget Luna) and Claude Fable 5.1 (new flagship, alongside Opus 4.8).

What it actually is

Upload documents, PDFs, slide decks, web links, or audio — Gemini Notebook builds a notebook that only answers questions grounded in those sources, with an inline citation back to the source passage for every claim. It also generates Audio Overviews (a podcast-style AI discussion of your sources) and Video Overviews, both genuinely useful for turning a dense source pile into something skimmable. The rename didn't change the core mechanic — it's still a source-grounded research tool, not a general chatbot.

Pricing as of September 2026

TierPriceNotebooks / sourcesDaily overviews
Free$0100 notebooks · 50 sources each3 audio + 3 video
Plus (Google AI Plus)$4.99/mo200 notebooks · 100 sources each6 audio + 6 video
Pro (Google AI Pro)$19.99/mo500 notebooks · 300 sources each20 audio + 20 video
Ultra$99.99–$199.99/mo500–600 sources100–200 audio + video

All tiers get the same features — Audio/Video Overviews, Deep Research — paid plans raise usage caps rather than unlocking new capability. Free tier requires no card and doesn't expire.

What it does well

Grounded, cited answers

Every claim traces back to a specific source and passage. For research where you need to defend a claim's provenance — academic work, legal research, competitive analysis — this citation discipline is the entire value proposition.

Audio Overviews

The AI-generated "podcast discussion" of your sources is a genuinely good format for reviewing a large source pile passively — on a commute, while cooking, wherever reading isn't practical.

Source-only discipline

Unlike a general chatbot, it won't quietly blend in outside knowledge or hallucinate beyond what your sources actually say. That constraint is a feature when accuracy matters more than breadth.

What it doesn't do well

General-purpose chat

It's not a ChatGPT or Claude replacement. No coding, no image generation, minimal help outside its source-grounded lane.

Real-time web research

You bring the sources; it doesn't independently search the live web the way Perplexity does. For "what's happening right now" queries, use Perplexity instead and feed the results in as a source if you want them grounded.

Long-form original writing

It synthesizes and summarizes your sources well; it's not built for drafting long original documents from scratch.

Who should use it

Researchers, students, analysts: Yes. Turning a stack of PDFs into a queryable, cited knowledge base is exactly what this tool is for.

Teams onboarding to a large internal doc set: Yes. Upload the docs, let new hires ask questions instead of reading everything cold.

Anyone needing live web search: Pair it with Perplexity — different jobs.

Casual chatbot users: Free tier is generous, but this isn't your daily-driver assistant — that's ChatGPT or Claude.

Bottom line

The rename to Gemini Notebook is cosmetic; the product underneath is still the best source-grounded research notebook available, free at a very usable tier. If your work involves interrogating your own document set with citations you can trust, this belongs in your stack alongside a general assistant and a web-search tool.