Self-hosted AI platform for legal teams. Docker Compose deploy, BYO model, matter-scoped chat with citation-verifiable answers. Apache 2.0.
LQ.AI is an open source alternative to Harvey and Legora. Self-hosted AI platform for legal teams. Docker Compose deploy, BYO model, matter-scoped chat with citation-verifiable answers. Apache 2.0. Apache-2.0 licensed, 107 GitHub stars, self-host with Docker, and last released 2026-07-04.
Should you switch?
Small law firms and in-house legal teams that want a self-hosted alternative to Harvey or Legora: matter-scoped AI chat, citation-verifiable answers, 17+ jurisdiction skill library, and a Docker Compose deploy that keeps data on your own infrastructure.
You want a managed SaaS with zero infrastructure to operate. LQ.AI is self-hosted only: you run Docker Compose, supply a Postgres instance, and bring your own LLM API key (Anthropic, OpenAI, or a local Ollama endpoint). No hosted trial exists as of 2026-07-17.
The project launched publicly in mid-2026 and has 107 stars as of 2026-07-17. Community is real and practitioner-built, but the user base is small. Treat it as early-stage infrastructure: evaluate before committing client data to it.
In the product
Community proof
Some quotes here endorse the approach rather than report first-hand use, or report a problem. We label those.
"As a trademark attorney using LQ.AI for IP matters, I want the platform to look up EU trademark register entries (via the EUIPO Trademark Search API) as a citable authority source, so that I can get live, verified answers about EU trademark registrations (status, owner, filing date, Nice classes) instead of relying on the model's training data or a separate manual search."
"For in-house practitioners, a simplistic litigation analysis skillset and would expand upon the LQ-AI in-house commercial focus."
Maintenance health
Can you do it today?
docker compose up -dGitHub
Support the creator
No public funding page yet.
More in Legal AI
See all in Legal AI ->