DEV.co
RAG Frameworks · theaiautomators

insights-lm-public

InsightsLM is an open-source, self-hosted alternative to Google NotebookLM that lets you chat with your documents, generate audio summaries, and ground AI responses in your own data sources. Built with React, Supabase, and N8N, it's designed for organizations wanting private, customizable document intelligence without vendor lock-in.

Source: GitHub — github.com/theaiautomators/insights-lm-public
646
GitHub stars
254
Forks
TypeScript
Primary language
MIT
License (OSI-approved)

Key facts

Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.

FieldValue
Repositorytheaiautomators/insights-lm-public
Ownertheaiautomators
Primary languageTypeScript
LicenseMIT — OSI-approved
Stars646
Forks254
Open issues4
Latest releasev0.2.0 (2026-01-15)
Last updated2026-01-16
Sourcehttps://github.com/theaiautomators/insights-lm-public

What insights-lm-public is

A TypeScript-based RAG application using React + Vite frontend (shadcn-ui, Tailwind), Supabase for authentication/storage/database, and N8N for workflow orchestration. Integrates with OpenAI/Gemini APIs for LLM and audio generation; separate fully-local variant uses Ollama/Qwen3. Deployed via edge functions with webhook-triggered N8N workflows.

Quickstart

Get the insights-lm-public source

Clone the repository and explore it locally.

terminalbash
git clone https://github.com/theaiautomators/insights-lm-public.gitcd insights-lm-public# follow the project's README for install & configuration

Need it deployed, integrated, or customized instead? DEV.co ships production installs.

Best use cases

Private Knowledge Base Chat

Organizations handling sensitive documents (legal, medical, financial) that need on-premise or self-hosted AI Q&A without sending data to external AI vendors.

Research & Document Analysis at Scale

Teams needing to rapidly extract insights and generate summaries from large document collections with verifiable source citations for audit/compliance.

Custom Internal Knowledge Tools

Companies wanting to white-label or extend document intelligence features into existing workflows without commercial licensing constraints (MIT allows this).

Implementation considerations

  • Setup involves ~6 steps spanning GitHub templating, Bolt.new import, Supabase migration, N8N workflow config, and secret management—plan 2–4 hours for first deployment.
  • Edge functions now handle auth internally with 'Verify JWT' flag disabled; ensure latest function code from repo is deployed to avoid security misconfigs.
  • N8N workflows can be auto-created via import JSON or manually configured node-by-node; auto-import recommended but requires paying Bolt account for migration execution.
  • Requires active API keys for OpenAI and Gemini (or equivalent LLM provider); cost scales with document volume and chat frequency.
  • Document upload/storage lives in Supabase; configure storage policies and CORS if exposing to external users.

When to avoid it — and what to weigh

  • Zero DevOps Tolerance — Setup requires creating Supabase projects, N8N instances, GitHub repos, and Bolt.new accounts. Not a plug-and-play SaaS; self-hosting demands infrastructure knowledge.
  • Fully Air-Gapped Environments — This variant (v0.2.0) requires OpenAI/Gemini API keys. For fully offline use, you must adopt the separate 'insights-lm-local-package' variant instead.
  • Production-Grade SLA Expectations — Project is 7 months old (created June 2025, latest release Jan 2026). Community-driven with 4 open issues. No commercial support agreement or guaranteed uptime.
  • Hands-Off Maintenance Preference — Supabase recently changed auth systems (anon/service role keys now 'legacy'). README warns you must use latest edge functions. Requires staying current with upstream changes.

License & commercial use

MIT License. Allows commercial use, modification, distribution, and private use with attribution. No copyleft obligations; you can fork, close-source, and commercialize derivative products. No warranty or liability protection.

MIT is OSI-compliant and permissive for commercial use. You can build and sell products on top of InsightsLM without royalties or licensing fees. However: (1) retain original MIT notice and copyright, (2) this repo itself carries no warranty—you assume all risk for production deployments, (3) any external service dependencies (OpenAI, Gemini, Supabase) have their own commercial terms that apply separately, (4) no trademark rights granted; 'InsightsLM' branding may be restricted. Review legal counsel before rebranding or heavy commercialization.

DEV.co evaluation signals

Editorial assessment — not user reviews. Directional, with an explicit confidence level.

SignalAssessment
MaintenanceActive
DocumentationAdequate
License clarityClear
Deployment complexityHigh
DEV.co fitGood
Assessment confidenceHigh
Security considerations

Self-hosting gives you data control but shifts operational security burden to you. Key points: (1) Edge functions now handle auth internally (good—reduces legacy key exposure), but you must ensure latest function code is deployed, (2) N8N webhook secrets must be stored securely in Supabase; oversharing these secrets breaks isolation, (3) document storage in Supabase requires proper row-level security (RLS) policies to prevent cross-user data leaks—README does not detail RLS setup, (4) OpenAI/Gemini API keys will be stored in N8N; ensure N8N instance itself is access-controlled, (5) no mention of encryption at rest/in transit for stored documents—verify Supabase encryption settings match your compliance needs, (6) MIT license includes no liability; you own all security incidents. Conduct threat modeling before production use.

Alternatives to consider

Google NotebookLM

Fully managed SaaS with built-in podcast generation, no self-hosting overhead, but closed-source, data privacy concerns, and vendor lock-in. Suitable if you accept cloud hosting and Google's ToS.

LangChain + LlamaIndex (DIY Stack)

Open-source Python libraries for RAG pipelines; vastly more flexible but requires engineering effort to build UI, workflows, and deployment. Better if you need deep customization and have strong eng resources.

Perplexity or Claude Projects

Proprietary tools offering document analysis and AI chat with external hosting. Trade-off: no self-hosting or commercialization rights, but mature, well-documented, and production-ready with support.

Software development agency

Build on insights-lm-public with DEV.co software developers

InsightsLM gives you full control over your data and AI workflows. Start with our quick-start guide, watch the setup walkthrough, and customize for your business—no licensing fees.

Talk to DEV.co

Related open-source tools

Surfaced by semantic similarity across the DEV.co open-source index.

Related on DEV.co

Explore the category and the services that help you build with it.

insights-lm-public FAQ

Can I run this entirely on-premises without cloud services?
Not with this variant (v0.2.0)—it requires OpenAI/Gemini APIs. Use the 'insights-lm-local-package' repo instead, which bundles Ollama, Qwen3, Whisper, and CoquiTTS for fully local operation.
What are the licensing implications if I modify and resell this?
MIT allows it. Retain the MIT license notice, give credit to original authors, and you're free to commercialize. However, consult legal counsel regarding trademark use ('InsightsLM' branding) and any third-party service dependencies.
How much does it cost to run?
Supabase free tier covers small deployments; Bolt.new and N8N have free-tier options but may require paid plans for production. OpenAI/Gemini API calls scale with usage. No single source of truth in README; estimate $100–500/month for modest production load.
Is there a managed hosting option or professional support?
Not mentioned in repo. The AI Automators (creators) offer a community at theaiautomators.com, but no formal SLA or commercial support tier is documented. You are responsible for operations.

Custom software development services

DEV.co is a software development agency delivering custom software development services to companies building on open source. Our software developers and web developers design, integrate, and ship production systems — spanning web development, APIs, AI, data, and cloud. If insights-lm-public is part of your rag frameworks roadmap, our team can implement, customize, migrate, and maintain it.

Ready to Deploy Private Document Intelligence?

InsightsLM gives you full control over your data and AI workflows. Start with our quick-start guide, watch the setup walkthrough, and customize for your business—no licensing fees.