AI Agent Index
ByHeather MacAvelia·Independently reviewed·Last verified Jul 8, 2026

Google source-grounded AI research assistant. Free up to 50 sources per notebook; paid tiers from $7.99/mo via Google AI subscriptions. Audio Overviews, mind maps, and citation-backed Q&A on uploaded sources only.

How we scored it

Autonomy

3/5

Integrations

4/5

Pricing clarity

5/5

Evidence

5/5

Setup

5/5

The facts

NotebookLM is the source-grounded AI research assistant from Google, built on a Retrieval Augmented Generation architecture that grounds every response strictly in user-uploaded sources rather than the open web. Users upload PDFs, Google Docs, websites, YouTube videos (via transcript extraction), or pasted text. The platform then synthesizes that material into summaries, citation-backed Q&A, and Audio Overviews (two-host, podcast-style conversations). It also generates Video Overviews, mind maps, slide decks exportable to PPTX, infographics, data tables, quizzes, and flashcards. Every response links back to the exact passage in the source document. The product has 10M+ installs and 267K Play Store reviews at 4.8 stars as of Q3 2026. Best fit: students running focused literature reviews, researchers synthesizing dense document collections, executives onboarding on reference material, and enterprise teams building accessible internal knowledge bases. Native source inputs include Google Drive (direct import), Google Docs (direct import), YouTube transcript extraction, URLs and websites, PDFs, and pasted text. Google Classroom integration was added in March 2026 for teacher-student workflows with shared notebooks. Exports include PPTX and standard audio files for offline Audio Overview playback. Google Workspace Business Standard plans and above include NotebookLM Plus capabilities for all seats. No public API exists for programmatic or external-agent integration, and Google has not published an MCP server as of Q3 2026. Several research integrations are missing. There is no Zotero or Mendeley import, so PDFs must be exported and re-uploaded manually. There is also no Overleaf or LaTeX connector, no Scopus or Web of Science citation import, no Notion or Obsidian connector, and no Slack or Teams integration for notebook sharing. NotebookLM cannot be purchased as a standalone product. All tiers are bundled with Google AI subscriptions. Standard (free, no card required): 100 notebooks, 50 sources per notebook at 500K tokens each, standard Audio and Video Overview generation. NotebookLM in Plus: bundled with Google AI Plus at $7.99/month, 100 sources per notebook, 2x generation limits, early feature access. NotebookLM in Pro: bundled with Google AI Pro at $19.99/month, 300 sources per notebook, 5x generation limits, higher Gemini model access. NotebookLM in Ultra: bundled with Google AI Ultra at $99.99/month (20TB storage, 500 sources per notebook) or $200/month (30TB storage, 600 sources per notebook). US students access Google AI Pro at $9.99/month. Workspace Business Standard starts at $14/user/month and includes NotebookLM Plus for all seats. NotebookLM Enterprise via Google Cloud carries HIPAA, SOC 2 Type II, ISO 27001, GDPR, and CCPA compliance for regulated-data workflows, with signed Business Associate Agreements available for healthcare teams. NotebookLM is not suited for open-domain research requiring web synthesis: it cannot access real-time information, crawl external databases, or retrieve sources not explicitly uploaded. For live web citations and autonomous source gathering, Perplexity AI ($17/mo) or ChatGPT Deep Research ($20/mo) cover broader source territory without manual upload workflows. For systematic review teams requiring PRISMA-compliant screening and structured data extraction, Elicit ($49/mo) or Rayyan (free basic tier, from $4.99/mo) provide dedicated pipelines NotebookLM does not include. For programmatic research automation, there is no official NotebookLM API: the Gemini API at $1.50 per million input tokens for Gemini 3.5 Flash provides model access without the NotebookLM interface. HIPAA coverage applies only to NotebookLM Enterprise via Google Cloud with a signed BAA, not to Standard, Plus, or Pro consumer tiers. Google restructured NotebookLM pricing at Google I/O on May 19, 2026, introducing four named tiers (Standard, Plus at $7.99, Pro at $19.99, Ultra at $99.99/$200) and retiring the former $249.99 Ultra price. Google Classroom integration launched March 2026. NotebookLM Enterprise received HIPAA certification March 2025. The underlying model runs Gemini 3.5 Flash on Standard and lower paid tiers, with higher Gemini model access on Pro and Ultra. No public API and no MCP server have been released by Google. Google Play shows 267K reviews at 4.8 stars with 10M+ installs, and G2 shows 18 reviews at 4.8 stars, reflecting strong satisfaction primarily through consumer and academic channels. The privacy policy confirms compliance with GDPR, CCPA, SOC 2 Type II, ISO 27001, and HIPAA for Enterprise tiers, and Google does not use uploaded sources to train its models.

Pricing

freemium · $7.99

View pricing ↗

Segment

b2b

Setup

easy

Verified

Jul 8, 2026

Transparency

Public

Contract

Month-to-month

Data training

Not Trained

Autonomy

Human Required

Capabilities

deep-researchcitationsliterature-reviewweb-search

Pros & Limitations

Editorial assessment

Pros

  • Source-grounded RAG architecture refuses to extrapolate beyond uploaded material, dramatically reducing hallucination compared to general-purpose LLMs on document-specific queries where source fidelity matters.
  • Audio Overviews convert dense documents into engaging two-host podcast conversations: a standout format with no direct competitor among research assistant tools, validated by 267K Play Store reviews at 4.8 stars as of Q3 2026.
  • Generous free tier covers most individual research needs at no cost: 100 notebooks and 50 sources each with no payment or credit card required, lowering the barrier for students and early-career researchers globally.

Limitations

  • No official API exists: programmatic or bulk workflows require the Gemini API at token-based pricing ($1.50 per million input tokens) rather than the NotebookLM interface, ruling out AI-agent automation or MCP integrations without custom development.
  • Cannot bridge uploaded sources with external knowledge: questions about current developments or cross-source synthesis fail unless all relevant material has been manually uploaded first, making it unsuitable for real-time or open-web research.
  • HIPAA coverage applies to Enterprise only: consumer tiers (Standard, Plus, Pro) do not qualify for HIPAA-regulated workflows, requiring healthcare teams to procure NotebookLM Enterprise through Google Cloud with a signed Business Associate Agreement.

Technical Details

Deployment
webmobile
Model architectureGemini models with Retrieval Augmented Generation (source-grounded)
Avg setup timeunder 5 minutes (Google account required)
Autonomous rateSource-grounded RAG: materially lower hallucination than general LLMs on document Q&A; requires human source upload and query initiation
Integrations
Google DriveGoogle DocsYouTubeGoogle Slides
Security
SOC 2 Type IIISO 27001HIPAAGDPRCCPA

Similar agents

Rating

4.1/ 5

Editorial score

How we score this →

Recognition

Transparent PricingListed 2026
Embed these on your site →

Industries

B2BEnterpriseSaaS

Leave a review

Never displayed publicly.

Agent Stacks

See workflow stacks that feature NotebookLM.

Compare

Related Content

Is this your tool?

Claim this listing free to get your embeddable badges, a review collection link, and the Verified checkmark.

Claim listing →

Free · Every Two Weeks

AI Agent Price & Rating Tracker

Price changes, new agent launches, acquisitions, and rating updates across 330+ AI agents, verified against live vendor data every 14 days.

No spam. Unsubscribe anytime. We never share your email.