Wren AI
Open-source data agent with text-to-SQL, MCP, and governed context across 20+ sources.
💬 What is Wren AI?
Wren AI is an open-source data agent that translates natural-language questions into governed SQL across more than 20 database sources. It is built for data and analytics leaders, BI teams, and product teams that need a single governed context layer shared by both people and AI agents such as Claude, ChatGPT, and Gemini via MCP. Distinctive capabilities include a Git-synced semantic layer (MDL), row- and column-level access policy, a GenBI app builder that turns one prompt into interactive dashboards, and an embedded analytics SDK with signed JWTs. Deployment ranges from cloud to fully air-gapped on-prem, the platform is SOC 2 compliant, and commercial plans ship with unlimited users and usage-based pricing. Production deployments span regulated industries including manufacturing, banking, and healthcare.
Who it's for
Best for: Data and analytics leaders at mid-market and enterprise companies who need governed natural-language SQL access for both human and AI-agent consumers.
- Natural-language business intelligence queries against a governed semantic layer
- Embedded analytics inside customer-facing products (iframe, API, or MCP)
- Giving Claude, ChatGPT, Gemini, or custom agents governed data access via MCP
- Self-hosted BI for regulated industries such as manufacturing, banking, and healthcare
- Prompt-to-dashboard generation for executives and operational reviews
Pros & cons
- Open-source core (Canner/WrenAI) with 17K+ GitHub stars
- 20+ database connectors with no ETL or migration
- Native MCP server for Claude, ChatGPT, Gemini, and custom agents
- SOC 2 compliant
- Unlimited users on every paid plan (usage-based, not per-seat)
- Open-core split: GenBI Apps, agentic mode, Git Sync, and enterprise governance require the commercial product
- Specific list pricing is not published on the main pages; requires the pricing page or a demo for exact quotes
- Air-gapped and large enterprise deployments typically require a sales-led onboarding process
- Semantic layer is managed via Git (MDL), which adds workflow overhead for teams not already using Git-based data tooling
Key features
- Natural-language to SQL translation (text-to-SQL)
- 20+ database connectors including Snowflake, BigQuery, Databricks, Redshift, PostgreSQL, SQL Server, Oracle, MySQL, ClickHouse, Athena, Trino, and Starburst
- MCP server for Claude, ChatGPT, Gemini, and custom agents
- GenBI Apps: prompt-to-dashboard builder with drill-down, roll-up, and filters
- Embedded analytics SDK via iframe, API, or MCP with signed JWTs and per-app tokens
- Git-synced semantic layer (MDL) for versioned, reusable metric definitions
- Unified data policy with row- and column-level access control and full audit log
- Onboarding agent for data source setup
Pricing
Prices listed on the Wren AI pricing page: $179/mo · $559/mo
Source: getwren.ai pricing page. Price checked 2026-10-02. Confirm current plans before buying.
💬 Community sentiment (from public reviews)
No third-party reviews found yet. Grade reflects our 27% verification confidence.
💬 Wren AI user reviews (0)
Write a reviewNo reviews yet. Be the first to review!
Trust & safety
Badge for Wren AI
OptionalListing is free and never requires a link. If getwren.ai links back to this page, the listing earns Partner perks: a higher category rank, a Partner badge on its card, and an ad-free direct “Visit” link. Use any link attributes you like — rel="nofollow" is fine.
Frequently asked
What data sources does Wren AI support?
It connects to 20+ sources including Snowflake, BigQuery, Databricks, Redshift, PostgreSQL, SQL Server, Oracle, MySQL, ClickHouse, Athena, Trino, and Starburst, with no ETL or migration required.
Can AI agents like Claude and ChatGPT use Wren?
Yes. Wren exposes an MCP server so Claude, ChatGPT, Gemini, or your own agents can call it as a sub-agent and receive the same governed answer a human user would.
Is Wren AI open source?
The context engine (MDL, text-to-SQL, MCP, CLI, and 20+ connectors) is open source under github.com/Canner/WrenAI. GenBI Apps, agentic mode, Git Sync, and enterprise governance ship in the commercial product.
How is pricing structured?
Pricing is usage-based, not per-seat, and every paid plan includes unlimited users. A cost calculator and pricing page are available on the site.
Can Wren run on-prem?
Yes. Deployment options include Wren AI Cloud, customer VPC, and fully air-gapped on-prem.