# Wren AI > Wren AI is the governed data agent shared by people and AI agents. Teams use Wren's built-in GenBI agent for plain-language answers, dashboards, and GenBI Apps. AI agents (Claude, ChatGPT, Gemini, or a customer's own) call Wren as a sub-agent over MCP to query their databases. Both get the same governed answer from one context layer the customer owns and versions in git. The context engine is open source (github.com/Canner/WrenAI, 17K+ stars); the commercial platform runs that same engine and adds GenBI Apps, agentic mode, AI-assisted context preparation, the UI, enterprise governance, and support. Works across 20+ data sources, deployed in Wren AI Cloud, a private VPC, or fully air-gapped on-premise. Key facts: - Category: governed data agent for people and AI agents; GenBI (Generative BI) and agentic analytics; open context layer (context engine) - Open core: the context engine, MDL context modeling, core text-to-SQL pipeline, MCP server, CLI, and 20+ connectors are open source, aimed at individual developers and used through the CLI (no UI); the commercial platform, built for teams and enterprises, adds GenBI Apps, agentic mode, context preparation, the UI and dashboards, access control with user and group management, row- and column-level security, advanced security, and support - Business definitions live in MDL files in the customer's own git repo, so there is no vendor lock-in - Deployment: Wren AI Cloud (managed, 7-day free trial), VPC, or air-gapped self-hosted (Enterprise Plus, licensed by concurrent sessions with unlimited users) - LLM-agnostic (OpenAI, Anthropic, Google Gemini, local models) and agent-agnostic (Claude, ChatGPT, Gemini, Cursor, or any MCP client calls Wren as a sub-agent) ## Key pages - [Open-core strategy](https://getwren.ai/open-core): Why the engine is open source, what is commercial, and how open core removes black-box and lock-in risk versus cloud-locked AI analytics - [Product: GenBI agent and MCP](https://getwren.ai/genbi): Wren's GenBI agent for your team, and MCP access for your AI agents, on one governed context layer - [Pricing](https://getwren.ai/pricing): Cloud plans (7-day free trial) and self-hosted Enterprise Plus - [Wren AI vs. alternatives](https://getwren.ai/versus): Comparison with ChatGPT, Claude, Gemini, Tableau, Power BI, Looker, ThoughtSpot, Databricks Genie, Snowflake Cortex, and more - [On-premise GenBI](https://getwren.ai/on-premise): Sovereign, air-gapped deployment with local models - [Embedded GenBI](https://getwren.ai/embedded): White-label conversational analytics and APIs for your product - [Industry use cases](https://getwren.ai/industries): Manufacturing, banking and finance, healthcare, retail, media, automotive - [Blog](https://getwren.ai/blog): Engineering and product articles ## Documentation - [Full site index for LLMs](https://getwren.ai/llms-full.txt): Every page's title, description, and FAQs, plus the gallery and blog index, generated from site content - [Open-source docs](https://docs.getwren.ai/oss/introduction): Install and run the open-source engine - [Commercial plan docs](https://docs.getwren.ai/cp/overview): Wren AI Cloud and Enterprise Plus - [GitHub repository](https://github.com/Canner/WrenAI): Source code, releases (weekly), issues ## Community - [Discord](https://discord.gg/5DvshJqG8Z): 1,700+ members - [Public roadmap](https://wrenai.notion.site/Wren-AI-Cloud-Public-Roadmap-1ed92976d0bf80488fe3fb16359734a3) ## Pages ### Wren AI: Open-Source GenBI Data Agent for Teams & AI Agents URL: https://getwren.ai/ Ask your data in plain language. Wren AI is the open-source GenBI agent your team uses and your AI agents call over MCP, governed by one context layer. ### GenBI Agent: Text-to-SQL, Charts & MCP Access | Wren AI URL: https://getwren.ai/genbi Ask in plain language and get governed SQL, charts, and GenBI Apps, or call the same agent from Claude, ChatGPT, or your own agents over MCP. 20+ sources. ### Open-Source Context Engine for AI Agents | Wren AI URL: https://getwren.ai/open-core The Wren AI context engine is open source on GitHub; every commercial plan runs that same engine in our cloud or yours. No black box, no lock-in. Q: What exactly is open source, and what is commercial? A: Open source: the context engine for individual developers, used through the CLI and MCP server (no UI), with MDL modeling, a core text-to-SQL pipeline, and 20+ connectors. Commercial: the advanced AI harnesses (GenBI Apps, agentic mode, AI-assisted context preparation), the UI and dashboards, embedded APIs, access control with user and group management, row- and column-level security, advanced security, and support, delivered as Wren AI Cloud or self-hosted Enterprise Plus. Q: What happens if we stop paying? A: Your MDL files already live in your git repo, your data never left your warehouse, and the open-source engine keeps running without a license. You lose the commercial platform (GenBI Apps, agentic mode, the managed experience), not your context or your data. Q: Can we start on open source and move to a paid plan later? A: Yes, that's the designed path. The engine and MDL are the same across OSS, Cloud, and Enterprise Plus, so the models you build on open source carry over. Many teams prove value on OSS or the free Cloud tier before rolling out company-wide. Q: How do we know the open-source project won't be abandoned? A: Judge the cadence, not the promise: weekly public releases, a public roadmap, 17K+ GitHub stars, and 1,700+ Discord members growing 5× year over year. The engine is the foundation of every commercial plan, so keeping it healthy is the business. ### Pricing: Unlimited Users, Cloud & Self-Hosted | Wren AI URL: https://getwren.ai/pricing Every Wren AI plan includes unlimited users. Cloud is billed by usage credits with a 7-day free trial; self-hosted by concurrent sessions. Start a trial or get a quote. Q: [Cloud] Is there a free trial, and do I need a credit card? A: Yes. Essential Cloud and Enterprise Cloud both start with a 7-day free trial. A credit card is required, and you can cancel any time before the trial ends. Prefer to start without a card? Talk to our sales team and we'll set it up. Q: [Cloud] How are credits charged? A: Credits are deducted only when an action succeeds; failed or errored actions are free. Asking questions, follow-ups, charts and the spreadsheet view each consume credits per the credit pricing table. Web credits cover work in the Wren AI Cloud UI. API credits cover API requests that invoke Wren AI models (answers, summaries, charts); non-AI calls like metadata or CRUD are free. Beta features are currently free. Any future pricing will be announced well in advance. Q: [Cloud] Do unused credits roll over, and what if I run out? A: On monthly plans, unused credits roll over up to 2× your monthly allowance. On annual plans, credits reset at the start of each billing cycle. Monthly free credits expire at the end of each month. If you run out, buy additional credits anytime from the billing section. Q: [Cloud] Can I switch between monthly and annual billing? A: Yes. Monthly → annual takes effect immediately; the prorated charge is applied at the end of your current monthly cycle. Annual → monthly takes effect when your current annual period ends. Q: [Self-hosted] What is a concurrent session? A: A concurrent session is one LLM-driven request in progress, such as answering a question or producing a chart, summary, report or GenBI App, from start to output, whether from the UI, API, Slack, Teams or MCP. AI authoring helpers (Modeling AI Assistant, AI Advisor) don't count. As a rule of thumb, 1 concurrent session supports about 5 active users; size for peak simultaneous use, not total headcount. Contact sales for an Enterprise Plus quote. Q: [Self-hosted] Do API requests cost extra? A: No. Requests through the API, including Slack, custom apps and AI agents, count toward your concurrent session license just like the UI. There are no per-call charges to track. Q: [Self-hosted] When do I need Enterprise Plus? A: When your data can't leave your network. Enterprise Plus runs the full platform in your own cloud, on-premise or air-gapped, licensed by concurrent sessions instead of credits. It adds SSO & SCIM 2.0, the evaluation framework, thread tracing and support escalation, and is required for OEM/resell, partner-led, regulated and bundled deals. Q: [General] What type of support do you offer? A: Email tickets through our support portal during business hours. See our Service Level Agreement for details. ### GenBI Buyer's Guide: Wren AI vs. ChatGPT, Tableau & More URL: https://getwren.ai/versus Compare Wren AI with ChatGPT, Claude, Tableau, Power BI, Databricks Genie, Snowflake Cortex and more on governance, openness, deployment, and pricing. Q: How is Wren AI different from every other option? A: It is the only one that combines three things: one governed data agent shared by your people and your AI agents, a context layer you own as open-source files in git, and deployment anywhere, including air-gapped. Q: Do we have to send our data to a vendor cloud? A: No. Wren AI is open source and self-hostable, including on-premises and air-gapped. It queries live data where it lives. Q: How does Wren AI fit our AI agent strategy? A: Your agents call Wren as a governed data sub-agent over MCP, under the same access rules your team works under. It's the data layer for the agents you already run. Q: Should we build this ourselves? A: Wiring an LLM to a database takes a weekend. Making it trustworthy at company scale takes shared definitions, row-level security, audit trails and constant upkeep. Wren AI gives you that foundation as open source. Q: How fast do we see value? A: Wren AI connects to live data in place with no migration, so most teams are asking real questions in days. ### GenBI by Industry: Finance, Healthcare & More | Wren AI URL: https://getwren.ai/industries See how finance, healthcare, manufacturing, retail, and media teams, and their AI agents, get governed answers from one context layer with Wren AI. ### Embedded AI Analytics & White-Label GenBI | Wren AI URL: https://getwren.ai/embedded Embed conversational GenBI in your product: white-label Embedded Threads, AI APIs, MCP, Slack and Teams. One context layer governs every answer. ### Git Sync: Your Context Layer as Code | Wren AI URL: https://getwren.ai/git-sync Move your open-source Wren project to enterprise GenBI over plain git. Models, metrics, and AI knowledge stay reviewable YAML and Markdown in a repo you own. Q: What file formats does Wren Git Sync store? A: Your context model syncs as YAML (models//metadata.yml, relationships.yml, and cubes/ for pre-aggregations), and your knowledge base syncs as Markdown and YAML under a committed knowledge/ folder. The only JSON in a Wren project is target/mdl.json, compiled build output that isn't committed. Everything you author and review is plain, open, human-readable text. Q: Does Git Sync work with GitHub, GitLab, and Bitbucket? A: Yes. Git Sync uses standard git against a repository you own, so it works with GitHub, GitLab, Bitbucket, or any git remote you already run. There is no proprietary transport: after a one-time bind, git push and git pull are the entire interface. Q: Can I trigger CI/CD pipelines on model changes? A: Yes. Because Git Sync is ordinary git, a git push is a normal push event your existing CI can react to: run validation on a pull request, diff a metric definition before it merges, and promote from staging to production the same way you ship application code. On the Wren side, a push to the bound branch queues a deploy that makes the updated model queryable. Q: Is any reusable git credential stored on my machine? A: No. The durable key lives at ~/.wren/cloud.yml with 0600 permissions and is only ever used to buy short-lived tokens. What git presents on each operation is a fresh Ed25519 JWT that expires in 600 seconds and is never written to disk or reused, so a leaked token is worth ten minutes and nothing more. Q: Do I need to learn a new CLI to push changes? A: No. There is deliberately no wren push and no wren diff. You bind a directory to a cloud project once with wren cloud create or wren cloud link; from then on, plain git push and git pull do exactly what they always have. Q: Can I take my context layer with me if I leave? A: Yes. The repository is yours, and the files in it are open Wren MDL in YAML plus knowledge in Markdown and YAML. You can read them, diff them, back them up, feed them to another tool, or clone the whole thing at any time. Wren Cloud is where the model becomes queryable and collaborative, not where it is trapped. ### ChatGPT Data Agent for Your Databases | Wren AI URL: https://getwren.ai/chatgpt-data-agent Turn ChatGPT into a governed data agent. Wren AI answers from live data across 20+ databases, with the SQL, a chart, and your business definitions. Q: How do I add Wren AI to ChatGPT? A: Open Wren AI in the ChatGPT app directory, add it to ChatGPT, and connect it to your Wren AI project. If you don't have a project yet, create one in Wren AI and connect a data source first. Then ask your first question in plain language. Q: How is this different from ChatGPT's built-in Data agent? A: A data agent that reads a warehouse directly has to infer what tables and columns mean. Wren AI adds the layer that settles it: metric definitions, declared joins, and business rules, written once in a git-versioned context model. The same definitions then serve ChatGPT, Claude, Slack, Teams, and your own agents, so the answers agree. Wren AI also reaches 20+ sources, including operational databases such as PostgreSQL, SQL Server, Oracle, and MySQL. Q: Does Wren AI answer from live data or from an export? A: Live data. Wren AI queries your connected sources when you ask, so the answer reflects what is in the database now. There is no ETL step and no data migration. Q: Who can see which rows and columns? A: Row-level and column-level security are enforced in Wren Engine at query time, based on who is asking. The rules you already set apply no matter which chat window the question came from. Read how Wren AI enforces role-based access control on AI-generated SQL. Q: Which databases does it work with? A: 20+ sources, including PostgreSQL, SQL Server, Oracle, MySQL, Snowflake, BigQuery, Databricks, Redshift, ClickHouse, Athena, Trino, Starburst. One Wren AI project can span several of them, and the engine plans each query against the source where the data already lives. Q: Can I try it before connecting production data? A: Yes. Wren AI core is open source on GitHub, so you can inspect the engine and the context model and run them against a sample database first. ### Claude Dashboards on Governed Data | Wren AI URL: https://getwren.ai/claude-dashboards Build Claude dashboards your CEO can trust. Claude's sub-agents work every angle of your data at once through Wren AI, with SQL behind every number. Q: How do I connect Claude to Wren AI? A: Enable the MCP connection on your Wren AI project, then search "Wren AI" in Claude's connector directory and authorize it with OAuth. A database credential never reaches Claude. The connector announcement covers setup in detail. Q: How do sub-agents build one dashboard together? A: Claude breaks a request like "build an executive dashboard on our expansion plan" into angles, such as trade areas, store ramp-up, formats, and operations, and works them in parallel. Each sub-agent queries through Wren AI, so all of them use the same metric definitions, joins, and access rules. Claude then merges the results, cross-checks the headline numbers, and builds the dashboard with each figure linked to its SQL. Q: Which databases can a Claude dashboard pull from? A: Every source Wren AI supports, 20+ in all, including PostgreSQL, SQL Server, Oracle, MySQL, Snowflake, BigQuery, Databricks, Redshift, ClickHouse, Athena, Trino, Starburst. One MCP connection reaches all of them, and the Wren AI engine plans each query against the source where the data already lives. There is no ETL step and no migration. Q: Can I trace a number on the dashboard back to its query? A: Yes. Every query runs through Wren AI, and every number on the dashboard links to the SQL that produced it. Each request is also logged with who asked and the SQL that ran. Q: Do two people asking for the same dashboard see the same data? A: Only if they are allowed to. Row- and column-level security apply at query time to the person asking, so each viewer sees only the rows and columns they are permitted to see. These controls and audit logs are part of Wren AI's paid plans. Q: What does Claude read before it writes SQL? A: Your context layer: what each column means, how tables join, and the business rules your team defined. These live in MDL as git-versioned files in your own repository, so a definition changed in a pull request is what Claude uses on its next question. Q: Does this only work in Claude? A: No. The same context layer serves ChatGPT, Slack, Microsoft Teams, and any MCP client through the Wren AI MCP server. Wren AI core is open source, so you can run the whole path locally first. ### MCP Server for Governed Data Access | Wren AI URL: https://getwren.ai/mcp-server Connect Claude, ChatGPT or your own agent to 20+ databases through one MCP server. Agents get governed SQL and answers under each user's permissions. Q: What is the Wren AI MCP server? A: It is a remote MCP endpoint, https://cloud.getwren.ai/api/mcp, that lets an AI agent hand data questions to Wren AI. The agent calls tools like ask or generate_sql, and Wren AI answers from live data through your context layer, so every caller works from the same business definitions. Q: Which agents and clients can connect? A: Setup is documented for Claude (web and desktop), ChatGPT and Claude Code. The server speaks the open Model Context Protocol over HTTP with OAuth, so your own agents and other MCP clients that support remote servers can use the same endpoint. See the MCP setup guide. Q: How does authentication work? A: Each person signs in with their own Wren AI account over OAuth. There is no API key to paste into a config file, and database credentials never leave Wren AI. Each connection is bound to one organization, and an admin can switch off a project or the whole MCP connection at any time, which cuts access immediately. Q: Who can see which rows and columns? A: Row- and column-level security is enforced on the server for the user who signed in, and the client cannot override it. Agents only reach projects an admin has exposed to MCP and the user is a member of. Because permissions follow the account, each person should connect with their own account rather than a shared one. Q: How is this different from a database MCP server? A: A database MCP server gives an agent a connection and a schema, so the model guesses what columns mean and how tables join on every question. Wren AI gives the agent your context layer instead: metrics, relationships and business rules defined once in MDL and versioned in git. One endpoint covers 20+ data sources, and every answer comes back with the SQL that ran. Q: Should my agent use ask or generate_sql? A: Use ask to delegate the whole question: Wren AI plans the query, runs it and answers, and may ask a clarifying question you reply to with respond_clarification. Use generate_sql, run_sql and generate_chart when your agent wants to control each step. Both paths resolve through the same context layer. Q: Which plan includes MCP? A: The hosted MCP server is included on every plan, starting with Essential Cloud. Agent calls draw from the same credits as your team (concurrent sessions on Enterprise Plus), and row- and column-level data controls apply to MCP callers on Enterprise Cloud and above. If you run the open-source Wren Engine, it ships its own MCP server you can host yourself; see the Wren Engine quickstart. ### Ask Your Data in Slack | Wren AI URL: https://getwren.ai/slack Mention @Wren AI in any Slack channel and get answers from live data in the thread, with charts and reasoning, on one governed context layer. Q: How do I add Wren AI to Slack? A: In Wren AI, open Settings, then Integrations, and click Add to Slack. Approve the permissions and your workspace appears on the integration page. Then choose a default project so Wren AI knows which data to answer from. The Slack setup guide walks through each step. Q: How do people ask questions? A: Mention @Wren AI in a channel, or send it a direct message without a mention. Questions can be in any language. Wren AI answers in a thread, and follow-up questions in that thread keep the context of the conversation. Q: Which data can each channel see? A: Questions in a channel use the default project unless you map that channel to a specific project. Mapping a channel such as #finance to a finance project keeps each team's questions on the data meant for them, and limits which data sources that channel can reach. Q: Can I see a chart, or how it got the answer? A: Yes. Reply with "Generate chart" in the thread to post a chart of the result, and open See reasoning to check how Wren AI read the question, including its assumptions and steps. Q: Does Slack apply row-level security per user? A: Not yet. Row-level security for Slack is on the roadmap. Today, control what a channel can reach by mapping it to a project that only contains the data that channel should see. For per-user row- and column-level security, use Wren AI directly or through the MCP server. Q: Will Slack answers match our dashboards and other tools? A: Slack asks the same context layer as every other Wren AI surface: metrics, joins and business rules defined once in MDL and versioned in git. Change a definition in a pull request and the next question in Slack uses it. Q: Which plans include Slack? A: Slack is available on the Essential Cloud plan with up to 3 mapped channels, and on Enterprise Cloud and above with unlimited channels. Each question and chart uses API credits like any other request. See pricing for details. ### Ask Your Data in Microsoft Teams | Wren AI URL: https://getwren.ai/microsoft-teams Mention @Wren AI in a Teams channel or group chat and get answers from live data in the thread, with charts and reasoning, on one governed context layer. Q: How do I add Wren AI to Microsoft Teams? A: Get the app from Microsoft Marketplace, or start in Wren AI under Settings, then Integrations, and click Add to Microsoft Teams. In Teams, click Sign in with WrenAI, pick your organization and a default project, and approve the Microsoft permissions. Setup needs a Microsoft 365 administrator and a Wren AI organization admin. The Teams setup guide covers each step. Q: Where can people ask questions? A: In channels and group chats, by mentioning @Wren AI, or in a direct message without a mention. Questions can be in any language. Wren AI answers in a thread, keeps the context for follow-ups, and suggests next questions you can send with one click. Q: Which data can each channel or group chat see? A: Conversations use the default project unless you map them. Under Channel and group to project mapping, add a mapping to point a channel or group chat at its own project, which also limits which data sources it can reach. Q: Can I see a chart, or how it got the answer? A: Yes. Each answer card has a Generate chart button, and View Thinking Process shows how Wren AI interpreted the question, including its assumptions and steps. Q: What permissions does it need, and is row-level security applied? A: Wren AI uses Microsoft Graph permissions for Teams and SharePoint to read the conversations it is added to and to reply. Row-level security per Teams user is not supported yet and is on the roadmap. Today, control what each conversation can reach through project mappings. For per-user row- and column-level security, use Wren AI directly or through the MCP server. Q: Will Teams answers match our other tools? A: Teams asks the same context layer as every other Wren AI surface: metrics, joins and business rules defined once in MDL and versioned in git. Change a definition in a pull request and the next question in Teams uses it. Q: Which plans include Teams? A: Teams is available on the Essential plan and above. Essential limits the number of mapped channels and group chats, and Enterprise allows unlimited mappings. Each question and chart uses API credits like any other request. See pricing for details. ### Sovereign AI Analytics: On-Premise GenBI | Wren AI URL: https://getwren.ai/on-premise Own your analytics AI. Run Wren AI on your own infrastructure with local models: data sovereignty by architecture, zero external AI calls, and fixed costs. Q: Which local models can Wren AI use? A: Wren AI runs against a curated set of open-weight models we test for SQL accuracy on GenBI workloads, served through any OpenAI-compatible endpoint such as Ollama or vLLM. We share the current validated list during your evaluation, and moving to a newer validated model is a configuration change, not a migration. Q: How does Wren AI address AI and data sovereignty requirements? A: Both are met by architecture rather than contract. Data sovereignty: every prompt, schema detail, and answer is processed inside your network and jurisdiction; nothing reaches an external AI vendor. AI sovereignty: inference runs on validated open-weight models on hardware you control, so the capability can't be repriced, deprecated, or revoked from outside. This holds in your data center, a sovereign-cloud tenancy, or a fully air-gapped site. Q: Can Wren AI run fully air-gapped? A: Yes. The context layer, the GenBI agent, and the model all run inside your network, so the deployment works with no internet route at all. On-prem and air-gapped deployment support is included in the Enterprise Plus plan. Q: Do local models produce good SQL? A: The quality of GenBI answers depends heavily on the model's own intelligence, so we maintain a list of tested, highly recommended local models and strongly suggest running one of them. Every request is also grounded in your context layer (schema, joins, metrics, and permissions), which narrows the model's job. Reach out to us to request the current list. Q: What does large-scale GenBI cost on-premise? A: Two fixed parts: inference runs on hardware you provision, so the cost per question falls as usage grows, and the Wren AI license is priced by concurrent sessions rather than seats, with unlimited users. Enterprise Plus is sized to your team, so contact sales for a quotation; estimate your sessions with the calculator or compare plans on the pricing page. Q: What is a concurrent session? A: A concurrent session is a single LLM-driven request to Wren AI (a question, chart, report, or GenBI App), counted from request to output, whether it comes from the UI, API, Teams, Slack, or MCP. Licensing counts simultaneous active sessions, not named users; as a guide, one concurrent session supports around five active users. Q: What is the fastest way to try it? A: Request a demo. We scope your deployment, share the validated local-model list, and set up a pilot in your environment on the Enterprise Plus plan, so what you evaluate is exactly what you roll out. ### Wren AI Cloud vs. Self-Hosted: Which Deployment Fits? URL: https://getwren.ai/cloud-vs-self-hosted Compare Wren AI Cloud and Self-Hosted side by side: where it runs, data boundary, AI models, pricing model, compliance, and which deployment fits your team. Q: Is it the same product on Cloud and Self-Hosted? A: Yes. Both run the same Wren AI GenBI platform: the same context layer, UI, dashboards, and governance. What changes is where it runs, how it reaches your AI models, and how it's licensed. Q: How is Self-Hosted different from the open-source version? A: The open-source project gives developers the context engine and CLI to build their own agents. The commercial self-hosted plan, Enterprise Plus, adds the full Wren AI platform on top: UI, dashboards, governance, security, and support. See the open-core page for what's included. Q: Do I need my own AI models or API keys? A: On Cloud, no: frontier models are managed for you and included in your plan's credits. Self-Hosted brings your own models: commercial APIs, or local and open-weight models running entirely on your hardware. See on-premise deployment for the local-model setup. Q: How is each option priced? A: Cloud is priced as monthly or annual plans with usage credits, with a 7-day free trial. Self-Hosted is an annual license sized by concurrent sessions rather than seat count. Use the session calculator to estimate what your team needs, and see pricing for both. Q: Is Wren AI Cloud SOC 2 compliant? What about air-gapped environments? A: Wren AI Cloud is included within the scope of our SOC 2 Type II examination; eligible customers can request the report under confidentiality terms via our security policy. For environments where data can't cross the network boundary at all, air-gapped deployment is supported on the Enterprise Plus self-hosted plan. Q: Can we start on Cloud and move to Self-Hosted later? A: Both deployments run the same platform, so starting on Cloud to prove value and moving to Self-Hosted as governance requirements grow is a common path. Talk to us and we'll help you plan the transition. ### Security & Governance for AI Analytics | Wren AI URL: https://getwren.ai/security How Wren AI governs every question from your team and AI agents: context-layer policy, row- and column-level security, MCP controls, and audit logs. ### Wren AI Demos: GenBI, MCP & Embedded Analytics Videos URL: https://getwren.ai/demos Watch short Wren AI demos: the GenBI agent answering in plain language, GenBI Apps from one prompt, Claude calling Wren over MCP, and embedded analytics. ### GenBI, Text-to-SQL, and Agentic Analytics | Wren AI URL: https://getwren.ai/blog Explore practical Wren AI guides on GenBI, text-to-SQL, context layers, AI agents, embedded analytics, and secure self-hosted deployment for data teams. ### Gallery: Analytics Apps Built from One Prompt | Wren AI URL: https://getwren.ai/gallery Browse dashboards and reports the Wren AI agent built from a single prompt against a governed context layer. See the prompt, the steps, and the artifact. ### Self-Hosted GenBI Pricing Calculator | Wren AI URL: https://getwren.ai/calculator Size a self-hosted Wren AI deployment in 60 seconds. Unlimited users and AI agents, licensed by concurrent sessions, not headcount. Get an exact quote. ### Phison × Wren AI: Air-Gapped GenBI with Private LLMs URL: https://getwren.ai/phison Wren AI's governed GenBI agent on the Phison AI Data Platform appliance: private LLMs on your own hardware. Fully air-gapped, with zero data egress. ### Wren AI Partner Program: Build a GenBI Practice URL: https://getwren.ai/partner Resell and implement Wren AI, the open-source GenBI agent. Elite Partners get qualified leads, hands-on enablement, co-marketing, and early access to new agent capabilities. ### Request a 30-Minute GenBI Demo on Your Data | Wren AI URL: https://getwren.ai/request-demo Request a 30-minute Wren AI demo for your data stack, governance needs, and AI-agent workflows. See the right deployment and leave with a clear next step. ### Contact Sales, Support, or Partnership Teams | Wren AI URL: https://getwren.ai/contact Contact Wren AI about Cloud, self-hosted deployment, OSS, security, support, or partnerships. Tell us what you need and we'll route it correctly. ### Support Portal for Cloud & Self-Hosted Customers | Wren AI URL: https://getwren.ai/support Submit a support request for Wren AI Cloud or Self-Hosted plans and our team will respond per SLA. Open-source users get help in Discord and GitHub Issues. ### Generative BI Comparison Guide: 15 Platforms | Wren AI URL: https://getwren.ai/download/comparison-guide Free guide: 15 GenBI platforms compared on 22 buyer-grade factors, the four differences that set Wren AI apart, the full tally, and a rollout blueprint. ### Embedded GenBI Build vs. Buy ROI Worksheet | Wren AI URL: https://getwren.ai/download/embedded-roi Free two-page ROI worksheet: what an in-house text-to-SQL stack really costs, the build vs. buy comparison, and a line-item calculator for your numbers. ### On-Premise GenBI Reference Architecture (PDF) | Wren AI URL: https://getwren.ai/download/on-premise-reference-architecture Get the complete blueprint for sovereign GenBI: governed, natural-language analytics inside your network with local models, a context layer, and fixed-cost TCO. ### Phison × Wren AI On-Prem GenBI Appliance Whitepaper URL: https://getwren.ai/download/phison-reference-architecture Co-branded reference architecture: aiDAPTIV+ acceleration, a governed agentic GenBI layer, air-gapped security, fixed-cost economics, 7-day rollout. ### Privacy Policy | Wren AI URL: https://getwren.ai/privacy-policy How Wren AI collects, uses, stores, and protects personal information across getwren.ai, Wren AI Cloud, and self-hosted deployments, and your rights. ### Security Policy | Wren AI URL: https://getwren.ai/security-policy Wren AI security policy: how we protect customer data, handle vulnerability reports, and operate Wren AI Cloud and self-hosted deployments securely. ### Service-Level Agreement | Wren AI URL: https://getwren.ai/sla Read Wren AI's Service-Level Agreement: uptime commitments, support response time targets, and service credit remedies for Wren AI Cloud customers. ### Terms of Use | Wren AI URL: https://getwren.ai/eula Read the Wren AI Terms of Use and Commercial Software License governing Wren AI Cloud, self-hosted deployments, and your use of getwren.ai. ### Wren AI vs. ChatGPT: The Governed Data Agent ChatGPT Calls URL: https://getwren.ai/versus/chatgpt ChatGPT's data agent borrows definitions from a semantic layer you maintain elsewhere. Wren AI is the governed, open-source data agent ChatGPT calls over MCP. Q: Can't we just point ChatGPT at our database? A: You can, and with the Data plugin it works well for teams already on Business or Enterprise: queries inherit the warehouse's row and column restrictions, and Enterprise adds audit logs. Metric definitions, though, come from a semantic layer you maintain elsewhere, and the whole setup lives inside OpenAI's cloud. Wren AI is that layer: open-source, self-hostable, with every answer traceable to SQL, and available to ChatGPT through MCP. Q: ChatGPT has warehouse plugins and a Data agent now. Isn't that enough? A: For many questions, yes. The Data agent can even read context from dbt, Snowflake Horizon or Databricks Genie. What it doesn't do is provide a governed layer of its own that other tools and agents share. Wren AI gives you one definition set that ChatGPT, Claude, Slack, Teams and embedded apps all resolve through, and you can run it on your own infrastructure. Q: Does Wren AI just wrap a model like ChatGPT? A: No. Wren AI orchestrates LLMs, but the difference is grounding: answers must resolve through your governed model and trace back to SQL. It's also open-source, self-hostable and warehouse-agnostic, so your context and configuration stay files you own rather than living in one vendor's chat product. ### Wren AI vs. Claude: The Governed Data Agent Claude Calls URL: https://getwren.ai/versus/claude Claude reaches your warehouse through connectors. Wren AI is the governed data agent Claude calls over MCP and your team asks directly, with traceable answers. Q: Claude already supports MCP and warehouse connectors. Why do I need Wren AI? A: Claude is a first-class MCP client with verified warehouse connectors that respect your existing roles. What those connectors don't give you is a shared, versioned definition of your metrics that also serves Slack, Teams, embedded apps and other agents. Wren AI is that governed MCP server, listed in the Claude Directory, so every answer resolves through one model and traces back to SQL. Q: Is Wren AI a competitor to Claude? A: No, they're complementary. Claude is the agent; Wren AI is the governed context layer it queries. Add Wren AI from the Claude Directory, authorize with OAuth, and you keep Claude's reasoning while answers stay grounded, secured per user, and auditable. Q: Can business users get value without using Claude? A: Yes. Wren AI is a full product: conversational analytics, GenBI Apps, and delivery in Slack, Microsoft Teams and your own apps for non-technical users, while exposing the same governed layer to Claude and other agents via MCP. One foundation serves humans and agents the same trusted answers. ### Wren AI vs. Gemini: Any Agent, Any Database, Beyond Google URL: https://getwren.ai/versus/gemini Gemini's data agents shine inside Google Cloud. Wren AI is one governed, open-source data agent for your team and your AI agents on any data source. Q: We're on Google Cloud and BigQuery. Why add Wren AI? A: Gemini's data agents are strongest when your model lives in Looker or BigQuery. If you want the context layer itself to be open-source, self-hosted, git-versioned and warehouse-agnostic, Wren AI fits, including as a governed layer over BigQuery alongside the other 20+ sources you own. Q: How is Wren's text-to-SQL different from Conversational Analytics in BigQuery or Looker? A: Both generate SQL and ground answers in a semantic model. Google's model is LookML or BigQuery-bound and its agents run in Google Cloud; Wren AI's context layer is open-source, git-native and self-hostable, works the same across 20+ sources, and traces every answer back to SQL so business users can trust the number. Q: Does Wren AI replace Gemini? A: It doesn't have to. Wren AI can govern BigQuery as one of its sources and add cross-source reach plus open, portable config-as-code. Many teams keep BigQuery and use Wren AI as the layer humans and agents query, especially to reduce dependence on a single cloud. ### Wren AI vs. Tableau: An Open-Source Tableau AI Alternative URL: https://getwren.ai/versus/tableau Tableau's AI agents live on the Salesforce stack. Wren AI is an open-source, warehouse-neutral data agent for your team and AI agents, no per-seat fees. Q: We've invested heavily in Tableau. Does Wren AI replace it? A: Not necessarily. Tableau remains excellent for pixel-perfect, curated dashboards. Wren AI answers ad-hoc questions in plain language and generates governed GenBI Apps in one prompt, no analyst ticket, no field-picking. Many teams keep Tableau for canonical reports and add Wren AI for self-serve. Q: Tableau has Tableau Agent, Pulse and Tableau Next now. How is Wren different? A: Tableau Agent, Pulse and the Tableau Next agents answer questions and build vizzes, with an MCP server for Cloud and Server. The richest pieces (Tableau Semantics, Concierge, Inspector) require Tableau Next on Salesforce Data 360 and a Cloud+ or Tableau+ subscription. Wren AI is open-source and warehouse-neutral: the context layer, skills and memory are files you own, every answer shows its SQL, and there is no per-seat pricing. Q: Do business users still need an analyst with Wren AI? A: No. Tableau now offers Agent Q&A on published data sources, but modeling and dashboards are still analyst work. Wren AI lets anyone ask in plain language and returns a governed answer with the SQL shown, and builds the dashboard from the same prompt. That's the core difference: Tableau is report-first, Wren AI answers the question directly. ### Wren AI vs. Power BI: A Copilot Alternative on Any Warehouse URL: https://getwren.ai/versus/power-bi Power BI Copilot needs Fabric capacity and stays in Microsoft's stack. Wren AI is a governed, open-source data agent on any warehouse, with no per-seat pricing. Q: Power BI is cheap inside Microsoft. Why pay for Wren AI? A: Power BI's seat price is low, but Copilot requires paid Fabric capacity (F2 or higher), and the model, DAX and tooling stay inside the Microsoft stack. Wren AI is ecosystem-neutral and open-source, priced by concurrent session with unlimited users, and works on any warehouse rather than only the Microsoft estate. Q: How does Wren AI compare to Copilot in Power BI? A: Copilot answers questions from your semantic model, builds report pages and, with Fabric IQ, shows its reasoning steps and DAX. It needs Fabric capacity and stays inside the Microsoft stack. Wren AI runs on any warehouse, is open-source, shows the SQL for every answer, adds reusable skills and memory, and versions the model, skills and memory as files. Q: Does Wren AI replace Power BI? A: Not necessarily. Power BI can stay your reporting layer while Wren AI adds open, agentic, cross-source GenBI and a governed context layer any agent can query via MCP, without the Microsoft ecosystem dependency or per-seat licensing. ### Wren AI vs. Looker: An Open Data Agent on Any Source URL: https://getwren.ai/versus/looker Looker's LookML is proprietary and quote-only. Wren AI is an open data agent for your team and AI agents, on a context layer you own, with no per-seat pricing. Q: We already model in LookML. Why switch to Wren AI? A: Looker pioneered the governed semantic layer, but LookML is proprietary, tied to one connection per model, and priced as a quote-only platform fee plus per-user licenses. Wren AI keeps the governed-model idea and makes it open-source, source-agnostic and agent-native, with model, skills and memory as versioned files agents can read and write. Q: Looker is already git-native and governed. What does Wren AI add? A: Agentic reasoning that compounds with reusable skills and memory, GenBI Apps in one prompt, a native MCP endpoint on every deployment (Looker's managed MCP server is in preview and not available on customer-hosted instances), 20+ sources behind one layer instead of one connection per model, and unlimited users with no per-seat pricing. Q: Can Wren AI sit alongside Looker or Gemini in Looker? A: Yes. Wren AI can govern the same warehouse Looker points at and serve humans and agents an open, portable layer, useful when you want to unify sources beyond a single connection or reduce dependence on the Google ecosystem. ### Wren AI vs. Amazon QuickSight: Any Agent, Any Cloud URL: https://getwren.ai/versus/amazon-quicksight QuickSight's AI is seat-based and AWS-only. Wren AI is a governed, open-source data agent on any warehouse and cloud, self-hostable, with no per-seat licenses. Q: We're on AWS with Quick Sight. Why add Wren AI? A: Quick Sight's base readers can be billed per session, but Amazon Quick chat, the generative layer, sits behind per-user Author Pro ($40/user/mo) and Reader Pro ($20/user/mo) licenses plus a $250/mo account fee, and the service is hosted only on AWS. Wren AI is ecosystem-neutral and open-source, with the same generative analytics across 20+ sources and any cloud, self-hostable, and unlimited users with no per-seat licenses. Q: How is Wren AI different from Amazon Quick chat? A: Both let anyone ask in plain language, and both ground answers in a governed layer (Wren AI's MDL; Quick Sight's Topics) with the generated SQL shown. Wren AI's differences are openness and portability: open-source, git-native configuration, a native MCP server any agent can call, and deployment on any cloud or on-prem. Q: Isn't Quick Sight cheaper than Wren AI? A: It depends on what you switch on. Base Quick Sight readers are inexpensive, but Amazon Quick chat, the capability worth comparing, requires Author Pro ($40/user/mo) and Reader Pro ($20/user/mo) seats plus a $250/mo account fee, or question-capacity bundles sold in blocks. Wren AI charges no per-seat licenses: unlimited users, priced by concurrent session on self-host, and it runs on any warehouse and cloud rather than only AWS. ### Wren AI vs. ThoughtSpot: An Open Data Agent You Own URL: https://getwren.ai/versus/thoughtspot ThoughtSpot Spotter is strong but proprietary and priced per user. Wren AI is an open-source, git-native data agent with transparent, usage-based pricing. Q: ThoughtSpot's Spotter is strong. Why choose Wren AI? A: Spotter 3 is a capable agentic analyst, and Wren AI matches that experience while staying open-source, git-native, and free of per-user licensing, with a native MCP endpoint in the open-source product (ThoughtSpot's is an Enterprise or Embedded add-on) and 20+ sources behind one governed layer. Q: ThoughtSpot is already warehouse-agnostic and git-native with TML. What's the difference? A: True, ThoughtSpot is strong on both. The remaining gaps are openness and economics: Wren AI is open-source and fully inspectable rather than proprietary, self-hostable including air-gapped, and priced by concurrent session with unlimited users instead of per-user plans ($25–50/user/month, Enterprise quoted). Q: Does Wren AI replace ThoughtSpot? A: It can, or sit alongside it. Both give business users ask-first analytics bound to a governed model; Wren AI adds open-source portability, config-as-code you fully own, and an MCP layer any agent can call. ### Wren AI vs. Metabase: A Data Agent, Not Just Dashboards URL: https://getwren.ai/versus/metabase Metabase is easy open-source BI. Wren AI adds a governed data agent for your team and AI agents: skills, memory, and natural-language SQL across 20+ sources. Q: We already run Metabase. Does Wren AI replace it? A: Not necessarily. Metabase is excellent for low-friction dashboards and self-serve queries, and many teams keep it. Wren AI adds a natural-language, agentic layer over a governed context model, with answers traceable to SQL and an MCP endpoint that exposes the model itself, across 20+ sources rather than one database at a time. They can run side by side, or Wren AI can take over the ask-a-question workflow. Q: Metabase is already open-source. What does Wren AI add? A: Metabase is open-core: Metabot and the MCP server are in the open edition, while row/column security, serialization and git sync are Pro and Enterprise features. Wren AI is open-source and agent-native, with natural-language-to-SQL, reusable skills, persistent memory and a governed MDL layer in the open product, plus self-host and air-gapped deployments and pricing that never charges per seat. Q: How is Wren AI's natural language different from Metabot? A: Metabot answers questions inside Metabase, shows its SQL, and starts each chat fresh. Wren AI is agent-first: it reasons in steps, grounds every answer in the governed model, remembers corrections across threads, and works identically across any connected source, keeping the model, skills and memory as version-controlled files you own. ### Wren AI vs. Apache Superset: A Governed Data Agent on Top URL: https://getwren.ai/versus/apache-superset Superset is mature open-source BI with no built-in assistant. Wren AI adds a governed data agent your team asks in plain language and AI agents call over MCP. Q: We run Superset today. Does Wren AI replace it? A: It can, or sit alongside it. Superset is great for SQL-fluent analysts building dashboards and exploring in SQL Lab. Wren AI adds the conversational, agentic layer on top, so non-technical users self-serve in plain language while answers stay grounded in a governed model and traceable to SQL. Many teams keep Superset for analyst exploration and use Wren AI as the ask-a-question front door. Q: Both are open-source. Why choose Wren AI? A: Superset is fully Apache-licensed and a favorite of data engineers, and Wren AI keeps the same open, self-hostable economics without per-user pricing. The difference is intelligence and audience: Superset's MCP server lets external AI clients query datasets, but each client brings its own model and context. Wren AI is natural-language and agent-first with a governed context layer shared by humans and every agent, so business users, not just SQL writers, get trusted answers. Q: Does Superset have anything like Wren AI's agent? A: Not as a built-in agent. Superset 6.1 ships an opt-in MCP server so tools like Claude Desktop can list datasets, run SQL, and generate charts and dashboards, and an in-app assistant is proposed but not merged; Preset offers a chatbot on its Enterprise plan. Wren AI brings a multi-step agent with reusable skills and persistent memory, grounds answers in a governed model, and shows the SQL behind every result, all open-source and portable across whichever sources you connect. ### Wren AI vs. Databricks Genie: Any Agent, Any Warehouse URL: https://getwren.ai/versus/databricks-genie Databricks Genie is powerful on Databricks. Wren AI is a governed, open-source data agent on any warehouse, self-hostable, with a git-versioned context layer. Q: Does Wren AI replace Databricks Genie? A: It does not have to. Wren AI can use Databricks as one of its sources and add cross-source reach, open-source portability, and a context layer that is not tied to a single platform. Many teams keep Databricks as their warehouse and use Wren AI as the open layer that people and agents query across every system they own. Q: We run everything on Databricks. Is Wren still worth it? A: Often, yes. Genie works well inside Databricks, and Lakehouse Federation can reach external databases read-only through Unity Catalog. Most organizations still have data that never lands in Databricks, such as an application Postgres, a finance system or a second warehouse. Wren AI connects to 20+ sources behind one governed layer, is open-source and self-hostable, and is priced by concurrent session with unlimited users, whereas Genie bills LLM usage per user plus separate SQL warehouse compute. Q: How does Wren's agent compare to Genie Agents? A: Both run multi-step, agentic workflows that turn natural language into SQL, and both now offer skills and memory (Genie's are in preview). The difference is reach and ownership: Wren AI runs the same way on any warehouse rather than only on Databricks, is open-source, and stores its model, reusable skills and memory as version-controlled files you can host yourself. Genie Agents can be defined in Databricks bundles, but they run only on Databricks SQL and Unity Catalog. ### Wren AI vs. Snowflake Cortex: Any Agent, Any Warehouse URL: https://getwren.ai/versus/snowflake-cortex Snowflake Cortex Agents run only on Snowflake data. Wren AI is a governed, open-source data agent for your team and AI agents on any warehouse, self-hostable. Q: Does Wren AI replace Snowflake Cortex? A: It does not have to. Wren AI can sit on top of Snowflake as one of its sources and add cross-source reach, open-source portability, and config-as-code that a single-platform tool can't provide. Many teams keep Snowflake as a warehouse and use Wren AI as the layer humans and agents actually query. Q: We are all-in on Snowflake. Is Wren still worth it? A: Yes. Snowflake CoWork is capable, but it reasons over data Snowflake stores or catalogs (native and Iceberg tables); other systems must be ingested first, and almost no company keeps every relevant number in one platform. Wren AI connects to the rest behind one governed, open layer, stays portable, and prices by concurrent session with unlimited users rather than per-token platform consumption. Q: How is Wren's text-to-SQL different from Cortex Analyst? A: Both turn natural language into governed SQL, and both offer agents, skills and MCP access. The difference is reach and ownership: Wren works identically across 20+ sources rather than only Snowflake-resident data, is open-source and self-hostable, and keeps its model, skills and memory as version-controlled files you own. ### Wren AI vs. Hex: An Open Data Agent You Own URL: https://getwren.ai/versus/hex Hex is proprietary, cloud-only, and priced per editor seat plus AI credits. Wren AI is an open-source, self-hostable data agent with unlimited users. Q: Hex has Threads and a Notebook Agent now. How is Wren AI different? A: Hex's agents are real and well built, and Threads now serves non-technical users directly. They live inside a proprietary, managed-cloud platform priced per seat plus AI credits. Wren AI is open-source and self-hostable, every answer traces to SQL against a governed model, and there are no editor or Explorer seats. Q: Hex also has semantic models and an MCP server. Doesn't that match Wren AI? A: On the checklist, Hex is closer than most. The difference is ownership: Hex's semantic models are authored in-app or synced from dbt, Cube or Snowflake, context can be synced from git on Team+ plans, its MCP server is Team and Enterprise only, and there is no customer-hosted option. Wren AI keeps the model, skills and memory as version-controlled files you own, exposes MCP on every deployment, and runs wherever you need it, including air-gapped. Q: Is Wren AI cheaper than Hex for a whole company? A: Usually, once business users enter the picture. Hex prices per editor ($36–75/editor/month) plus separately paid Explorer seats on Team and Enterprise plans, metered AI credits, and embedding and single-tenant as Enterprise add-ons. Wren AI never charges per seat: unlimited users, with self-host priced by concurrent session, so rolling ask-a-question analytics out to hundreds of stakeholders doesn't multiply a license. ### Wren AI vs. Glean: The Data Agent Behind the Number URL: https://getwren.ai/versus/glean Glean answers questions over docs and apps and hands warehouse queries to others. Wren AI is the open, governed data agent that owns the number. Q: Glean can query Snowflake, Databricks and BigQuery now. Isn't that the same as Wren AI? A: Not quite. Glean routes the question to Snowflake Cortex, Databricks Genie or BigQuery under each user's own credentials and combines the result with document context. Glean itself has no metric model, so consistency depends on the semantic views or Genie spaces your data team maintains. Wren AI is the governed layer across 20+ sources, with every answer traceable to SQL. Q: Should we buy Glean or Wren AI? A: They do different jobs. Glean answers knowledge questions over documents and SaaS apps; Wren AI answers data questions over your warehouses with governed metrics. Many enterprises will run both, and because both speak MCP, Glean's agents can call Wren AI as their governed analytics tool. Q: Glean is enterprise-grade. Is Wren AI? A: Yes: SOC 2 Type II, role-based access, and every answer auditable back to SQL and a versioned model. The structural difference is control: Wren AI is open-source with true self-host and air-gapped deployment, while Glean offers SaaS or Customer Hosted (Glean-managed software running in your GCP or AWS account, which you cannot patch or modify yourself) with quote-only pricing, either per user or by FlexCredit consumption. ### Wren AI vs. Kore.ai: The Data Sub-Agent Kore Agents Call URL: https://getwren.ai/versus/kore-ai Kore.ai is an agent platform with no warehouse querying or metrics. Wren AI is the governed data agent its agents call over MCP, and your team uses directly. Q: We're standardizing on Kore.ai for agents. Where does Wren AI fit? A: As the analytics layer those agents call. Kore.ai's platform is an MCP client, and Wren AI ships a native MCP server, so a Kore-built agent can hand data questions to Wren and get back governed answers grounded in your warehouse, with the SQL to prove them, rather than answering from indexed content. Q: Can't Kore.ai's search answer data questions? A: From content it has indexed or actions it calls. Its knowledge bases cover SaaS and document sources like SharePoint and Confluence, and "structured data" there means uploaded CSV, XLSX or JSON, not live SQL against your warehouse. There's no text-to-SQL engine, no semantic layer for metrics, and no warehouse connectors. Wren AI generates governed SQL against live data across 20+ sources. Q: Is Wren AI an agent platform too? A: Wren AI ships its own multi-step analytics agent with reusable skills and memory, but it isn't trying to run your contact center or IVR. It's the open, governed context layer any agent platform (Kore.ai included) queries for trustworthy numbers. Complementary categories: Kore.ai orchestrates the agents, Wren AI owns the analytics answer. ## Gallery - [Customer Risk and Transactions Analysis](https://getwren.ai/gallery/customer-risk-transactions): A risk desk view of 2,000 banking customers and 157,224 transactions: high-risk segmentation, transaction value, and failed-transaction rate, with a drilldown tab per dimension. (Banking; Risk & compliance) - [Banking Customer and Transaction Analysis](https://getwren.ai/gallery/banking-customer-transactions): Eleven widgets over customer behavior, card portfolio mix, transaction patterns, geographic concentration, and risk segmentation: time series, category mix, geo, heatmap, radar, bubble, and dual-axis views. (Banking; Customer analytics) - [Merchant Analysis Report](https://getwren.ai/gallery/merchant-analysis): The full transaction history read by merchant category: transaction mix, customer risk segments, card brands, and top geographies across Jan 2022 to Oct 2024. (Banking; Revenue & growth) - [eCommerce Business Metrics](https://getwren.ai/gallery/ecommerce-business-metrics): An analyst-style operating readout of completed-order sales, customer concentration, product demand, inventory risk, web activity, and geographic revenue concentration. (E-commerce; Executive overview) - [Customer Health Dashboard](https://getwren.ai/gallery/customer-health): A CS and RevOps read on account health: average and total MRR, churn rate, and the large neutral middle whose sentiment hides early churn risk. (E-commerce; Customer analytics) - [Churn & Retention](https://getwren.ai/gallery/churn-retention): A five-part diagnostic of where revenue is leaking: logo churn, gross MRR churn, NRR and GRR, average tenure at churn, and which accounts deserve intervention first. (E-commerce; Customer analytics) - [Inventory Management](https://getwren.ai/gallery/inventory-management): Where inventory capital is parked, how demand compares with supply, where flow is breaking down, and which SKUs need attention next, with turnover, weeks of cover, and reorder signals. (E-commerce; Operations) - [Customer Purchasing Behavior](https://getwren.ai/gallery/customer-purchasing-behavior): The path from global demand to category-level product opportunity, weighing momentum, penetration, repeat behavior, and basket adjacency to pick the strongest category candidate. (E-commerce; Product & marketing) - [Revenue and Churn Analysis](https://getwren.ai/gallery/saas-revenue-churn): A SaaS operating view across 3,000 accounts: current MRR base, active-account share, churn rate, and portfolio health score, split into overview, plans, usage, and retention tabs. (SaaS; Revenue & growth) - [Clinical Outcomes Performance Report](https://getwren.ai/gallery/clinical-outcomes): Whether care delivery is producing good outcomes across 3,500 encounters in 10 departments: a composite quality lens over mortality, readmission, complications, improvement, and patient satisfaction. (Healthcare; Operations) - [Executive Hospital Dashboard](https://getwren.ai/gallery/hospital-executive-view): An executive KPI rail (encounter volume, readmission rate, trend performance, and mix) presented with chart-level callouts so the important signals read without studying the charts. (Healthcare; Executive overview) - [Patient Experience Analysis Report](https://getwren.ai/gallery/patient-experience): Sixteen charts plus KPI widgets on patient experience, channel mix, financials, and risk, with an interactive department filter for drilling into a single service line. (Healthcare; Customer analytics) - [Diagnosis Report](https://getwren.ai/gallery/diagnosis-report): 5,531 diagnosis records across 969 patients and 3,500 encounters: primary diagnosis rate, present-on-admission rate, and more than ten widgets on volume, outcomes, demographics, and financial performance. (Healthcare; Operations) - [Patient Demographics & Care Settings](https://getwren.ai/gallery/patient-demographics): A population read on 1,000 patients and 3,500 encounters: age and gender balance, insurance mix, where care activity concentrates across outpatient and inpatient settings, and readmission burden. (Healthcare; Customer analytics) - [Oil & Energy Operations Report](https://getwren.ai/gallery/oil-energy-operations): Portfolio production and revenue against target (1.46bn barrels, $106.92bn revenue, 75.79% capacity utilization) with downtime, safety, and intensity metrics pointing at where operational focus should shift. (Oil & Gas; Operations) - [Distributor Delivery Performance Report](https://getwren.ai/gallery/distributor-delivery): 20.4K deliveries across 10 distributors, 49 pickup states, and 49 delivery states: 87.6% on-time rate, 90.3% successful-delivery rate, and a 68.5-hour average delivery time. (Manufacturing; Operations) - [Production Monitoring Dashboard](https://getwren.ai/gallery/production-monitoring): A plant executive layout with a left KPI rail and 20 widgets: 99.6% yield on 4.78M good units, production and labor cost trends, and top production lines by runtime versus downtime. (Manufacturing; Executive overview) - [Recruitment Intelligence Report](https://getwren.ai/gallery/hr-recruitment): A talent operations view of sourcing mix, weekly applications, pipeline conversion, hiring performance, recruiting spend, scorecard weights, and upcoming interviews. (Human Resources; Talent acquisition) - [Leave Management Intelligence Report](https://getwren.ai/gallery/leave-management-hr): A workforce report on leave demand, request outcomes, category mix, absenteeism, policy compliance, and departmental utilization from January through November 2025. (Human Resources; Workforce management) - [Student Admissions Intelligence Report](https://getwren.ai/gallery/student-admissions-report): A dense admissions analysis of 3,883 applicants across test participation, academic preparation, enrollment, housing, residency, demographics, and cohort profiles. (Education; Education analytics) - [University Room Utilization Report](https://getwren.ai/gallery/university-room-utilization): A Spring 2026 facilities report covering 104 rooms and 5,555 seats, with utilization by room type, capacity range, weekday, time slot, and building. (Education; Operations) ## Blog - 2026-10-10 [Wren AI Is Now Live in the ChatGPT App Directory](https://getwren.ai/post/wren-ai-chatgpt-app-directory): Wren AI brings conversational business intelligence into ChatGPT. Connect your company's databases, ask in plain language, and get answers from live data with the SQL, visuals, and context behind them. - 2026-10-10 [Claude Dashboards on Wren AI: One Question, 24 Governed Queries](https://getwren.ai/post/claude-dashboards-on-wren-ai): Claude can now build dashboards from a single prompt. Connected to Wren AI, it reads your context layer first, queries 20+ databases through one MCP connection, and every number on the dashboard links back to the SQL that produced it. - 2026-10-06 [Analytics Belongs Where the Work Happens: Four Embedded Demos You Can Try](https://getwren.ai/post/embedded-analytics-where-the-work-happens): Nobody leaves your product to go ask a dashboard a question. The question shows up mid-task, about the thing on screen. We built four working demos, in capital markets, retail, healthcare, and insurance, to show what happens when the answer lives there too. - 2026-10-04 [How Wren AI turns a question into an answer](https://getwren.ai/post/how-wren-ai-turns-a-question-into-an-answer): Ask in plain language. Wren retrieves the right business context, writes and validates the SQL, and runs it against your own database. - 2026-09-28 [How Wren AI Enforces Role-Based Access Control on AI-Generated SQL](https://getwren.ai/post/how-wren-ai-enforces-role-based-access-control): Wren AI doesn't replace your data security policy, it enforces it. Row-level and column-level security are applied in Wren Engine at query time, driven by who is asking, even when the SQL was written by an AI. - 2026-09-17 [The Agent Didn't Get Smarter. The Context Did.](https://getwren.ai/post/the-agent-didnt-get-smarter-the-context-did): At OpenAI's Data Agent launch, almost nobody asked about the agent. Every question was about the layer between the agent and the answer: context, governance, reproducibility. Here's why that layer, not the model, decides whether the number is right. - 2026-09-11 [Move your open-source Wren project to the cloud over plain git](https://getwren.ai/post/wren-cloud-git-sync): Git Sync moves your open-source Wren project, the MDL semantic model, and the knowledge that makes answers correct into Wren Cloud as plain YAML and Markdown files in a git repository you own. - 2026-09-01 [16× Cheaper, 37% Faster: Wren AI Cloud Now Defaults to GPT-5.6 Luna](https://getwren.ai/post/wren-ai-now-runs-on-gpt-5-6-luna-by-default): Wren AI Cloud is now 16× cheaper and 37% faster per task, with accuracy holding at 95% of evaluation assertions passed — GPT-5.6 Luna is the new default model. Here's what we measured before flipping the switch. - 2026-08-24 [Introducing the Wren AI Gallery](https://getwren.ai/post/introducing-the-wren-ai-gallery): The Gallery is a growing library of real dashboards and reports built by the Wren AI agent from a single prompt. Open any example to see the exact prompt and every step behind it, then point the same template at your own warehouse. - 2026-08-10 [Wren AI Is Now Available in Microsoft Teams](https://getwren.ai/post/wren-ai-microsoft-teams-connector): Wren AI is now listed on Microsoft Marketplace. Ask a question in plain language inside a Teams chat or channel and get the answer in-thread, with the SQL and a chart, resolved against the same context layer every other surface uses. - 2026-08-09 [Wren AI Is Now Available in the Claude Directory](https://getwren.ai/post/wren-ai-claude-directory-connector): Wren AI is now listed in the Claude Directory. Add the connector inside Claude, authorize with OAuth, and Claude asks questions of your context layer instead of your raw schema, with row-level security and a full audit trail at query time. - 2026-08-02 [The Dashboard Was Never the Asset](https://getwren.ai/post/the-dashboard-was-never-the-asset): For twenty years we treated the dashboard as the deliverable, but the durable asset was always underneath it: what your numbers actually mean. Dashboards depreciate. Context compounds. - 2026-07-25 [The Genie Meter Is On: What Databricks' New Pricing Says About the ROI of GenBI](https://getwren.ai/post/genie-meter-is-on-roi-of-owning-your-genbi): Databricks turned the Genie meter on, and agents get no free tier. Why full autonomy makes an agnostic GenBI layer, any model on any platform, the whole decision. - 2026-07-14 [You Can't Trust an AI Agent You Can't Debug.](https://getwren.ai/post/trust-ai-agent-thread-tracing-evaluation): An AI agent that answers business questions has to be debuggable and measurable, or 'earned trust' is just a slogan. How thread tracing, benchmarks, and the AI Advisor close the loop. - 2026-06-22 [GenBI for AI Agents: Why We Rebuilt Wren AI Around an Open Context Layer](https://getwren.ai/post/genbi-for-ai-agents-open-context-layer): The consumer of analytics is becoming the agent, and agents need your business's meaning in a form they can read. A founder's case for the open context layer, and why we rebuilt Wren AI OSS around it. - 2026-06-03 [Introducing Agentic GenBI: A Chat Box Answers a Question. An Agent Does the Work.](https://getwren.ai/post/introducing-agentic-genbi): Agentic GenBI moves BI past the chatbot: AI agents that reason in steps, remember context, and stay governed on an open context layer. - 2026-04-02 [The Missing Context Layer for AI Agents Over Business Data](https://getwren.ai/post/the-missing-context-layer-for-ai-agents-over-business-data): Why we rebuilt Wren Engine from a context layer into an open context engine, and what we learned along the way - 2026-01-21 [Zero Joins, Infinite Insights: 3 HR Use Cases for Wren AI](https://getwren.ai/post/zero-joins-infinite-insights-3-hr-use-cases-for-wren-ai): HR data is often trapped behind complex SQL joins and busy analyst schedules. Discover how Wren AI democratizes business intelligence by allowing anyone to query employee data using natural language. - 2026-01-14 [Say Hello to Interactive Mode: A New Way to Converse with Your Data](https://getwren.ai/post/wren-ai-interactive-mode-launch): Ad-hoc Mode is for speed; Interactive Mode is for depth. With our major new update, Wren AI now offers a truly conversational environment built for complex, multi-step workflows. - 2026-01-09 [The Future is a Data Fabric and Mesh, Woven by AI Agents](https://getwren.ai/post/the-future-is-a-data-fabric-and-mesh-woven-by-ai-agents): Why the Future of Enterprise Data Mesh and Fabric Relies on Autonomous AI Agents and a Universal Semantic Protocol. - 2026-01-06 [Your dbt Models Are Your Context Layer: Native dbt Integration in Wren AI](https://getwren.ai/post/no-more-modeling-start-asking-dbt-wren): Wren AI's new dbt integration is here. Connect your existing dbt project and turn your dbt models into your Wren AI context layer, no duplicate setup required. Your governed dbt models are instantly ready for natural language querying, accelerating your time-to-insight and democratizing access to trustworthy data across your organization. - 2025-12-21 [Wren AI 2025 Year in Review: From Open Source to Agentic BI in Production](https://getwren.ai/post/wren-ai-2025-year-in-review-from-open-source-to-agentic-bi-in-production): How we grew Wren AI to over 13,000 stars and over 10K users of Wren AI today, expanded query-in-place support across the modern data stack, and laid the foundation for agentic, semantic-driven analytics. - 2025-12-09 [Announcing Native Databricks Support in Wren AI](https://getwren.ai/post/announcing-native-databricks-support-in-wren-ai): Today, we’re excited to announce that Wren AI now integrates natively with Databricks, making it easier than ever for organizations to turn their lakehouse into a fully conversational analytics platform. - 2025-12-01 [The Analyst's Secret Weapon: How We Can Save a 13K Dollar Leak in Our Supply Chain In 17 Minutes](https://getwren.ai/post/save-13k-supply-chain-leak-17-minutes): How Wren AI transformed a passive BI role into a proactive investigation. Using AI rapidly uncovered that poor profit margins were caused by a high-defect supplier and a negative Net Contribution Per Unit for the high-volume Cosmetics line. This led to immediate decisions, including switching a shipping route to save over $13,000. - 2025-11-25 [The End of the Great Centralization: Why the Future of Enterprise Data is Distributed](https://getwren.ai/post/the-end-of-the-great-centralization-why-the-future-of-enterprise-data-is-distributed): How GenBI and AI Agents are replacing the costly “Single Source of Truth” model with a decentralized, pipeline-free SQL architecture. - 2025-11-14 [AI's Role in B2B Procurement Platform. How Embedded Analytics Elevates Sourcing, Pricing, and Strategy](https://getwren.ai/post/ai-role-b2b-procurement-embedded-analytics-sourcing-pricing-strategy): Today procurement is crippled by data overload. Wren AI acts as an embedded analytics layer, augmenting existing platforms with APIs and a semantic query layer to help teams source smarter, negotiate faster, and plan with confidence. - 2025-11-03 [From 80% Loss to 40% Profit: An AI-Powered Inventory Turnaround](https://getwren.ai/post/from-80-percent-loss-to-40-percent-profit-an-ai-powered-inventory-turnaround): Your inventory data holds secrets. We dove in with a Generative BI tool like Wren AI to uncover them. In minutes, we found a best-selling product that was actually a profit-killer, pinpointed the true source of our spoilage costs, and turned a money-losing promotion into a strategic tool to convert waste into profit. See the simple questions that unlocked thousands in savings. - 2025-10-24 [How We Closed the Profitability Gap by Talking to Our Logistics Data](https://getwren.ai/post/how-we-closed-the-profitability-gap-by-talking-to-our-logistics-data): Stop guessing. In a game of minutes and margins, the platform that understands its data best will win. Traditional dashboards only show you the problem; they don't give you the solution. See how teams are using conversational AI to move beyond static reports, pinpoint the real bottlenecks, and turn data into immediate, actionable strategy. - 2025-10-21 [Wren AI Launches Native dbt Integration for Governed, AI-Driven Insights](https://getwren.ai/post/wren-ai-launches-native-dbt-integration-for-governed-ai-driven-insights): We are excited to announce a major upgrade: Wren AI now fully supports dbt (data build tool)! Wren AI automatically syncs your models, descriptions, and relationships, turning your dbt investment into immediate, accurate GenBI with zero manual effort. - 2025-10-17 [Conversational Analytics on Hotel Data-Turning 50K Reviews into Insights with Wren AI](https://getwren.ai/post/how-wren-ai-turns-hotel-reviews-into-actionable-insights): No complex schemas. No SQL. Just your data - and instant answers. - 2025-09-30 [Stop Waiting for Reports: Get Ad-Hoc Patient Insights Instantly with Wren AI](https://getwren.ai/post/stop-waiting-for-reports-get-ad-hoc-patient-insights-instantly-with-wrenai): Transform complex medical data into immediate, actionable answers through simple conversational queries, no manual analysis required. - 2025-09-17 [Beyond Dashboards How Conversational AI is Revolutionizing Structured Finance Analytics](https://getwren.ai/post/beyond-dashboards-how-conversational-ai-is-revolutionizing-structured-finance-analytics): How Wren AI transforms complex financial data into instant, actionable insights through natural language, no technical expertise required - 2025-09-04 [Milliseconds to Millions Wren AI That Turns Ad Marketplace Chaos into Speed and Clarity](https://getwren.ai/post/milliseconds-to-millions-wren-ai-that-turns-ad-marketplace-chaos-into-speed-and-clarity): How Wren AI Transforms Ad Marketplaces with Real-Time Insights, Unified Metrics, and Natural Language Queries - 2025-08-27 [Wren AI vs. Vanna: The Enterprise Guide to Choosing a Text-to-SQL Solution](https://getwren.ai/post/wren-ai-vs-vanna-the-enterprise-guide-to-choosing-a-text-to-sql-solution): Choosing between a governed GenBI platform and a text-to-SQL component library - 2025-08-20 [Game On: How Wren AI Transforms Backend Gaming Data Into Actionable Insights](https://getwren.ai/post/games-on-unlocking-value-from-backend-gaming-platform-analytics): How Wren AI Turns Natural-Language Questions Into Cost-Efficient SQL to Accelerate Insights from Gaming Data - 2025-08-20 [How Wren AI Agent Supercharges the Future of Marketing Measurement Platforms](https://getwren.ai/post/how-wren-ai-agent-supercharges-the-future-of-marketing-measurement-platforms): Marketing Measurement Platforms (MMPs) are pivotal for performance marketers seeking to unify data and fuel more innovative campaign strategies. - 2025-07-31 [How Wren AI Revolutionizes Automotive Inventory Management for Non-Technical Users](https://getwren.ai/post/how-wrenai-revolutionizes-automotive-inventory-management-for-non-technical-users): Empowering dealerships with real-time, plain-English insights to accelerate sales, optimize inventory, and put data in every team member’s hands. - 2025-07-18 [Conversational BI Empowering Self-Service Analytics with Wren AI’s GenBI](https://getwren.ai/post/conversational-bi-empowering-self-service-analytics-with-wren-ais-genbi): Self-service analytics has reshaped business intelligence (BI) by enabling non-technical users (marketers, customer success managers, executives, and operations teams) to explore data independently, without relying on data experts. - 2025-07-08 [Unlock CRM Superpowers with Generative BI Real-Time Insights, Natural Language Queries, and Seamless API Integration](https://getwren.ai/post/unlock-crm-superpowers-with-generative-bi-real-time-insights-natural-language-queries-and-seamless-api-integration): Transform your CRM from static dashboards to dynamic, AI-powered analytics using Wren AI’s GenBI APIs, deliver instant, personalized insights without technical bottlenecks. - 2025-07-03 [Less Waiting, More Winning How Smart Marketing Agencies Use AI to Drive Results Faster](https://getwren.ai/post/less-waiting-more-winning-how-smart-marketing-agencies-use-ai-to-drive-results-faster): Replace slow BI dashboards with conversational analytics. Ask questions in plain English and get CTR, CPC, and ROAS metrics visualized in seconds. - 2025-06-26 [How AI-Powered Analytics Transformed a Brokerage’s Trading Platform](https://getwren.ai/post/how-ai-powered-analytics-transformed-a-brokerages-trading-platform): GenBI Delivers 10x Faster Data Insights for Crypto Trading Firms Using Natural Language Queries - 2025-06-09 [From Spreadsheets to GenBI Why Enterprise Analytics Needs Real-Time Database Access](https://getwren.ai/post/from-chatgpt-spreadsheets-to-live-sql-genbi-text-to-sql-why-enterprise-analytics-needs-real-time-database-access): The Ultimate Guide to Choosing Between ChatGPT’s File Upload Feature and Text-to-SQL Solutions for Data-Driven Decisions - 2025-04-21 [Powering AI-driven workflows with Wren Engine and Zapier via the Model Context Protocol (MCP)](https://getwren.ai/post/powering-ai-driven-workflows-with-wren-engine-and-zapier-via-the-model-context-protocol-mcp): Use semantic understanding, SQL generation, and email automation to streamline data requests with LLM agents and Wren’s powerful query engine. - 2025-04-11 [Powering Semantic SQL for AI Agents with Apache DataFusion](https://getwren.ai/post/powering-semantic-sql-for-ai-agents-with-apache-datafusion): Bridge the Gap Between AI and Enterprise Data with a High-Performance Semantic Layer and Unified SQL Interface for Model Context Protocol (MCP) - 2025-04-01 [Beyond Text-to-SQL: Why Feedback Loops and Memory Layers Are the Future of GenBI](https://getwren.ai/post/beyond-text-to-sql-why-feedback-loops-and-memory-layers-are-the-future-of-genbi): How Wren AI’s Innovative Approach to Question-SQL Pairs and Contextual Instructions Delivers 10x More Accurate Generative Business Intelligence - 2025-03-31 [Fueling the Next Wave of AI Agents Building the Foundation for Future MCP Clients and Enterprise Data Access](https://getwren.ai/post/fueling-the-next-wave-of-ai-agents-building-the-foundation-for-future-mcp-clients-and-enterprise-data-access): How Wren Engine Powers the Semantic Layer for AI-Driven Workflows, Unlocking Context-Aware Intelligence Across the MCP Ecosystem - 2025-03-18 [Transform Your Database Management with an AI-powered spreadsheet-like Interface](https://getwren.ai/post/transform-your-database-management-with-an-ai-powered-spreadsheet-like-interface): Operate your database effortlessly using natural language, no SQL required. - 2025-02-13 [From Traditional BI to GenBI Embracing a Smarter, More Human Approach](https://getwren.ai/post/from-traditional-bi-to-genbi-embracing-a-smarter-more-human-approach): How Generative BI Transforms Data Analytics with AI-Powered Insights and Human-Centric Decision Making - 2025-02-07 [AI-Powered Business Intelligence: The Complete Guide to GenBI](https://getwren.ai/post/the-future-of-business-intelligence-the-generative-bi-genbi): Discover how Generative Business Intelligence (GenBI) powered by Wren AI is transforming data access with conversational AI, real-time insights, and intuitive decision-making tools for modern enterprises - 2025-02-07 [What is GenBI? Get Your First Dashboard Done in 3 Clicks](https://getwren.ai/post/what-is-genbi-get-your-first-dashboard-done-in-3-clicks): GenBI: The Future of Business Intelligence, Create Data-Driven Dashboards in 3 Clicks - 2025-01-17 [Forward to 2025 Powering the Future of Enterprise with AI-Driven Data Intelligence](https://getwren.ai/post/forward-to-2025-powering-the-future-of-enterprise-with-ai-driven-data-intelligence): A Year of Wren AI Growth and Vision for the Future - 2025-01-07 [Reducing Hallucinations in Text-to-SQL Building Trust and Accuracy in Data Access](https://getwren.ai/post/reducing-hallucinations-in-text-to-sql-building-trust-and-accuracy-in-data-access): How Schema Grounding, Semantic Layers, and Iterative Validation Can Enhance Text-to-SQL Reliability - 2025-01-03 [How Uber is Saving 140,000 Hours Each Month Using Text-to-SQL, And How You Can Harness the Same Power](https://getwren.ai/post/how-uber-is-saving-140-000-hours-each-month-using-text-to-sql-and-how-you-can-harness-the-same-power): Discover how Uber’s Text-to-SQL technology streamlines data queries and learn how to apply this efficiency to your operations. - 2024-12-31 [Wren AI 2024 Year in Review A Fruitful Journey and Exciting Plans Ahead](https://getwren.ai/post/wren-ai-2024-year-in-review-a-fruitful-journey-and-exciting-plans-ahead): Reflecting on Wren AI’s 2024 Growth, Launching Wren AI OSS/Cloud, Expanding Data Sources, and Looking Ahead to 2025 - 2024-12-21 [Is Data Modeling Dead with Text-to-SQL? How Wren AI Bridges Modern BI and Traditional Practices](https://getwren.ai/post/is-data-modeling-dead-with-text-to-sql-how-wren-ai-bridges-modern-bi-and-traditional-practices): Discover why data modeling is more essential than ever in the age of Text-to-SQL solutions. Learn how Wren AI leverages well-structured, business-oriented datasets to deliver accurate, intuitive insights—combining the best of traditional BI best practices with cutting-edge natural language analytics. - 2024-12-12 [Trino + Text-to-SQL - Getting Big Data from Anywhere Fast and Easy with AI](https://getwren.ai/post/trino-text-to-sql-getting-big-data-from-anywhere-fast-and-easy-with-ai): How Trino and Wren AI Combine to Supercharge Big Data Analytics, Streamline Multi-Source Integration, and Deliver Real-Time Insights - 2024-12-10 [Why the Semantic Layer is Essential for Reliable Text-to-SQL and How Wren AI Brings it to Life](https://getwren.ai/post/why-the-semantic-layer-is-essential-for-reliable-text-to-sql-and-how-wren-ai-brings-it-to-life): Unlocking the Power of Natural Language Queries: How the Semantic Layer Transforms Text-to-SQL with Wren AI - 2024-11-03 [Celebrating Hacktoberfest 2024: A Heartfelt Thanks to Our Incredible Wren AI Contributors](https://getwren.ai/post/celebrating-hacktoberfest-2024-a-heartfelt-thanks-to-our-incredible-wren-ai-contributors): Celebrating Hacktoberfest 2024 with 11 new contributors and 15 merged pull requests! A huge thank you to all who helped enhance Wren AI. Excited for more collaboration in 2025 as we continue building and innovating together! - 2024-10-11 [Wren AI Hacktoberfest 2024 Starts Today!](https://getwren.ai/post/wren-ai-hacktoberfest-2024-starts-today): Hacktoberfest 2024 is here! We're inviting developers of all levels to join our open-source community. Together, we'll build Wren AI as a friendly community for all. - 2024-10-08 [Wren AI on Trino Broadcast Seamlessly Integrating AI with Trino](https://getwren.ai/post/wren-ai-on-trino-broadcast-seamlessly-integrating-ai-with-trino): Last week, our team was featured on the Trino Community Broadcast to discuss how to use Wren AI for chatting with Trino! - 2024-08-20 [Can AI Solve Data Silos Challenge? New Challenges To The Multi-AI Agents Era.](https://getwren.ai/post/can-ai-solve-data-silos-challenge-new-challenges-to-the-multi-ai-agents-era): In this post, we are going to discuss what businesses encounter data silo challenges before the LLM era and the new hope LLM brings to the data ecosystem, and finally, we will explore some of the new challenges that emerge when we are using more vertical LLM agents to solve different purpose of scenarios. - 2024-08-20 [How do we rewrite Wren AI LLM Service to support 1500+ concurrent users online?](https://getwren.ai/post/how-do-we-rewrite-wren-ai-llm-service-to-support-1500-concurrent-users-online): How Wren AI LLM Service leverages Haystack and Hamilton as performant Generative AI infrastructure - 2024-08-20 [How do you use LangChain to build a Text-to-SQL solution? What are the challenges? How to solve it?](https://getwren.ai/post/how-do-you-use-langchain-to-build-a-text-to-sql-solution-what-are-the-challenges-how-to-solve-it): In this post, using LangChain as an example of how you could use it to implement text-to-SQL, encounter challenges and how you can solve… - 2024-08-20 [How Snowflake Is Building the Most Powerful SQL LLM in the World](https://getwren.ai/post/how-snowflake-building-the-most-powerful-sql-llm-in-the-world): Last month, Vivek Raghunathan, VP of Engineering at Snowflake, discussed the pioneering Snowflake Copilot, here's what we learned. - 2024-08-20 [How we design our semantic engine for LLMs? The backbone of the context layer for LLM architecture.](https://getwren.ai/post/how-we-design-our-semantic-engine-for-llms-the-backbone-of-the-semantic-layer-for-llm-architecture): The advent of Trend AI agents has revolutionized the landscape of business intelligence and data management. In the near future, multiple… - 2024-08-20 [The new wave of Composable Data Systems and the Interface to LLM agents](https://getwren.ai/post/the-new-wave-of-composable-data-systems-and-the-interface-to-llm-agents): The new interface to bridge AI agents and the data system. How can we manage and design to meet the future demands of multi-AI agents? - 2024-08-16 [How to use OpenAI GPT-4o to query your database?](https://getwren.ai/post/how-to-use-openai-gpt-4o-to-query-your-database): Today, OpenAI released its latest LLM model, GPT-4o. People are sharing crazy applications built on top of this groundbreaking model. By… - 2024-07-02 [How to use Meta Llama 3 to query MySQL database using Ollama and Wren AI](https://getwren.ai/post/how-to-use-meta-llama-3-to-query-mysql-database-using-ollama-and-wren-ai): Step-by-step tutorial on hosting private LLM endpoints through Ollama and using the latest open model Meta Llama 3 to query your MySQL