Open-source business intelligence with Metabot, an AI assistant that answers questions, writes SQL and summarises dashboards on your own semantic layer and permissions.
Lightdash
Summary
Open-source BI that reads metrics straight from a dbt project, with AI agents that answer questions, build charts and dashboards, and maintain the semantic layer.
Screenshots
Description
Lightdash is an open-source business intelligence platform for teams that model their data in dbt. Instead of redefining metrics inside the BI tool, Lightdash reads dimensions and metrics from the dbt project, so analysts define them once in code and business users explore them through a point-and-click Explore view, charts and dashboards. Charts can also be saved as code and versioned alongside the dbt project.
AI agents sit at the centre of the current product. A team sets up an agent with its metrics and context, and people ask it questions in threads: compare segments and time periods, dig into the charts and tables that come back, and ask the agent to turn a finding into a new chart or dashboard in the same conversation. The agents learn from use, building what Lightdash calls a self-improving context layer, and Autopilot (October 2026) automates project maintenance in the background. Lightdash also exposes an MCP server so agents in other tools can query the same governed metrics, and supports embedding analytics into products.
The open-source edition is self-hosted and covers core BI features with community support on Slack. Cloud Pro is $3,000 per month with unlimited users and no per-seat pricing, and adds AI agents and MCP, the metrics catalog, custom roles, guided onboarding and a choice of cloud or on-premises deployment; there is a free trial.
Reviews
No reviews yet — sign in to write the first one
Similar App Suggestions
See all Lightdash alternativesA business intelligence platform built on a shared semantic model, with an AI chat that answers questions over governed data, plus dashboards, spreadsheets and SQL.
Embedded analytics for SaaS products: white-label dashboards, charts and an AI chat agent your customers use inside your app, Slack or Claude.
Open-source, OpenTelemetry-native observability platform for logs, metrics and traces — with first-class tracing for LLM calls and agent runs.
Open-source newsletter platform that bills for emails sent, not dormant subscribers, with an MCP server so AI agents can run it.