Open-source agentic GenBIgithub.com/Canner/WrenAI

Answers your team and AI agents can trust.

Ask in plain English and get governed SQL, charts, and GenBI apps across 20+ data sources, in our cloud or yours.

Try Wren AI Cloud free
Request a demo
17,260#1 GenBI on GitHub

Open source · 20+ data sources · Cloud or self-hosted

01Built for every team and leader

From one prompt to a live dashboard your leadership team can act on.

Ask in plain language and Wren AI builds a real-time dashboard on your governed data, with drill-down, roll-up and global filters built in.

Wren AI
HC
Expansion Capacity dashboard
✦

ASK WREN AI

|

Generate
Reading your prompt

Trusted by data teams worldwide

Why Wren AI / Open core

The engine is open source. That's the difference.

AI analytics platforms ask you to trust a black box with your data. Wren AI is open core: the context engine that answers your questions is public on GitHub, and your definitions live in your own git.

See the open-core strategyCompare the alternatives →
Wren AI platformCommercial · Teams & enterprise
GenBI Appsagentic modesecurity & governance+ open core included
Context engineMDL · text-to-SQL · MCP · CLI · 20+ connectors
Open source · Individuals
Your data · queried in place
02Why Wren AI

Context is the foundation of intelligence

Context is constructed automatically and flows into every answer, so humans and agents never drift from the sources of truth.

Agentic Design01

Not just chat. Sandboxed, multi-step reasoning.

Tools and scripts run inside an isolated environment.

Agents query, chart, extract from PDFs, build dashboards, save skills.

Every step is traceable and replayable.

Top customers this quarter?Wren AI reasoningUnderstand intentPlan trusted queryReturn answerRanked customers
Agent-Friendly Architecture04

Git-native + file system. MDL the agents read and write directly.

Skills, Memory, Semantic Model, and Instructions are all files: versioned, branched, PR-reviewed, rolled back.

Open MDL means one definition lives in MDL; agents reason over it instead of hallucinating joins.

20+ data sources: BigQuery, PostgreSQL, ClickHouse, Amazon Redshift, and dialect-agnostic.

Agent contextskillsmemorymodelSemantic modelCustomersOrdersProducts
Unified Data Policy05

Auditable by design; not as a workflow, but as infrastructure.

RLS and CLS enforced at query time, across every human or agent.

Role-based access with full activity logs. Provable who saw what, which query produced which number.

Multi-tenant cloud, private cloud, and air-gapped on-prem.

One policy applies whether the caller is the UI, the API, or an MCP client.

AppAPIMCPOne policyGoverned resultregionrevssn***Audit logallowedmaskedaudited
Explore the platform
03Embed Wren AI

Ship conversational analytics inside your product.

Embed white-label GenBI through threads, APIs, or MCP. One context layer governs every answer. Your brand, your policies, live in days.

Embedded Threads

iframe

The full conversational experience in your app with one snippet. Your logo, your color, and every answer scoped to the end user asking.

index.html
<iframe
  src="https://cloud.getwren.ai/iframe/id"
  width="100%" height="100%">
</iframe>

Embedded AI API

REST

Build your own UX on the raw endpoints: natural language to governed SQL with execution, plus chart specs you render in your stack.

terminal
curl -X POST '.../api/v1/generate_sql' \
  -d '{"question": "Top customers by LTV?"}'

# { "sql": "SELECT ...", "threadId": ... }

MCP

agents

Connect Claude or ChatGPT over OAuth. Agents query through your semantic definitions and never touch raw tables.

setup
# Claude & ChatGPT connect over OAuth
> "What drove revenue growth in Q2?"
  generating SQL... running... charting...
Governed by one context layerSigned JWTsRow & column-level scopingFull audit logsPer-app tokens
Explore embedded analyticsRequest a demo
04Use Cases

One context layer.
The same trusted answer for everyone.

Deliver an AI analytics strategy your executives will actually believe.

Ship "AI for analytics" without ripping out the warehouse or trusting a black box. Get a governed context layer, self-service in Slack and Teams, and a roadmap you can defend to the board.

Governed Context LayerON-PREMISEDatabasesSQL Server, Postgres, MySQLCLOUDWarehousesSnowflake, BigQueryMETADATAModelingdbt, data catalogsTeams & SlackSelf-service answersAI AgentsGoverned automationExecutive ReportsBoard-ready numbersDashboardsAlways-on metricsAccess ControlSemantic ModelData PoliciesKnowledge Center
swipe
05Case studies

How teams put governed GenBI to work

See how manufacturing, healthcare, finance, retail, and media teams use governed answers across the systems they already own.

01 / 05
06Proof

Results from teams in production

Used by data teams and public companies across the US, EU, Middle East, and Asia.

0x
Superhuman Speed

Agents query through a context layer that already knows your business. Time-to-answer in seconds.

0%
Less Waiting

Questions that used to sit in an analyst queue get governed answers in seconds, so teams stop waiting on tickets.

0+
Hours Saved per Month

Recurring reporting on autopilot, a five-person data team reclaims 100+ hours every month.

“Wren AI enables natural language data interaction across 20+ databases without ETL or migration. Integrated with Phison’s aiDAPTIV+ architecture, it delivers up to an 80% cache hit rate in the on-prem AI infrastructure and secure AI operations, accelerating adoption with lower costs and faster deployment.”
Wei
CTO, Phison (A Public Company)
“We use Wren AI for our SaaS product, DemandSense, to give our clients the ability to interact with their data naturally - asking questions, generating insights, and creating ad-hoc charts and dashboards in real time. It’s been a game-changer for making analytics more accessible and actionable without requiring technical expertise.”
Anna
CTO, Impactable
“Wren AI has transformed our workflow enabling our teams to get instant insights through natural language, without complex reporting processes. Our decision speed improved by over 50%, and data shifted from an "expert tool" to a "company-wide asset." Wren AI is more than analytics. It’s revolutionized our data culture.”
Shasta
CEO, Nextlink (A Public Company)
Choose your path

Start with the deployment that fits.

Try Wren AI Cloud free, build with the open-source context engine, or compare enterprise deployment options.

Try Wren AI Cloud freeExplore Wren AI OSSCompare plans
Wren AI