
ChatLab
Not yet source-checkedLocal-first analytics and AI agent for your personal chat history, with bring-your-own model support.
Machine-drafted from the vendor's site · not yet source-checkedProfile updated Link checked How we rate
In short
ChatLab is a free, open-source local tool that analyzes your personal chat history using your own AI models. It keeps data private on your machine while offering 30+ built-in analyses and natural-language querying.
Pick ChatLab if you want to mine years of personal chat history with AI without uploading it to a third-party cloud.
Skip it if you need a turnkey hosted chatbot or you don't want to wire up your own LLM and data exports.
ChatLab is a desktop and CLI application that ingests your exported chat history from mainstream messaging apps and turns it into a queryable, analyzable personal corpus. It ships with 30+ built-in analyses, including sentiment scoring, topic detection, activity heatmaps, relationship reviews, knowledge mining, session summaries, smart and deep search, and a conversation replay view. An AI Agent layer sits on top, letting you ask natural-language questions and run SQL queries against your own messages without uploading them to a third party.
The pitch is privacy: data stays in a local database on your machine, and the team explicitly recommends running local models when hardware allows. That makes it a fit for power users, journalists, researchers, and anyone who wants to mine years of WhatsApp/Telegram/WeChat-style history without handing it to a SaaS. A standardized data-abstraction layer plus custom adapters means you're not locked to one chat app. The project is open-source on GitHub with an npm-installable CLI (`chatlab-cli`), and the underlying AI is BYOM rather than tied to a specific provider.
The site doesn't publish pricing, model lists, or supported adapters in detail, and the AI quality will depend entirely on whichever model you wire up. Treat it as a power-user toolkit rather than a polished consumer product.
ChatLab is the rare AI tool that takes privacy seriously by default - your messages stay on disk and you bring the model. The 30+ canned analyses plus a SQL layer give it more depth than the usual 'chat with your data' wrapper, but the docs are thin on which messengers and models actually work today.
— The AI Tool Bible editorial team
Pros
- ✅ Local-first: chat data and settings never leave your machine
- ✅ 30+ built-in analyses plus an AI Agent and SQL query layer
- ✅ Open-source with both desktop app and npm-installable CLI
- ✅ Adapter model lets you plug in arbitrary chat platforms
Cons
- ⚠️ No published list of supported chat platforms or AI models
- ⚠️ AI quality depends entirely on whichever model you bring
- ⚠️ Power-user tool; setup is heavier than a hosted SaaS
Facts
- Platforms
- cli, api
- API available
- Yes
- Open source
- Yes (license: AGPL-3.0)
- GitHub stars
- 7,646 (checked Nov 13, 2026)
- Last GitHub push
- Nov 4, 2026
- First commit
- Nov 2025
- Model used
- Bring-your-own (local models recommended)
- Vendor pages
- Roadmap
- Package downloads (30 d)
- 1,145 · npm chatlab-cli · checked Nov 9, 2026
- Earliest Wayback capture
- 2025-12-29 (earliest archived capture of the site, not a launch date)
Where this tool sits
Use cases
Frequently asked
- How much does ChatLab cost?
- ChatLab is free and open-source. It uses a bring-your-own-model approach, meaning you provide the AI model, so there are no subscription fees or usage costs associated with the software itself.
- Which chat platforms does ChatLab support?
- It ingests exported history from mainstream messaging apps, including WhatsApp, Telegram, and WeChat. A standardized data-abstraction layer with custom adapters ensures you are not locked into a single specific chat application.
- Is my chat data uploaded to the cloud?
- No. ChatLab is local-first, meaning your data stays in a local database on your machine. The team explicitly recommends running local models to ensure your personal history is never uploaded to a third-party cloud.
- Do I need to configure my own AI model?
- Yes. ChatLab uses a bring-your-own-model (BYOM) system. You must wire up your own LLM, and the quality of AI responses depends entirely on the specific model you choose to install and use.
- What kind of analyses can I run?
- It includes 30+ built-in analyses such as sentiment scoring, topic detection, activity heatmaps, and relationship reviews. You can also ask natural-language questions or run SQL queries against your messages via the AI Agent layer.