[{"data":1,"prerenderedAt":596},["ShallowReactive",2],{"page-pt-BR-/compare/cube":3,"page-parent-pt-BR-compare":377},{"id":4,"title":5,"author":6,"authorImage":6,"authorTitle":6,"body":7,"category":6,"date":6,"description":352,"extension":353,"faq":354,"image":6,"meta":370,"navigation":371,"path":372,"readTime":6,"seo":373,"seoTitle":374,"stem":375,"__hash__":376},"content/compare/cube.md","vibedasher vs Cube",null,{"type":8,"value":9,"toc":343},"minimark",[10,14,19,23,27,35,269,273,289,293,316,322,326],[11,12,5],"h1",{"id":13},"vibedasher-vs-cube",[15,16,18],"h2",{"id":17},"who-cube-is-for","Who Cube is for",[20,21,22],"p",{},"Cube calls itself \"the agentic analytics platform built on a semantic layer\", serving \"both humans and AI agents — for your team and your customers\". The semantic model is the centre of the product: everything downstream — natural-language chat, SQL and visual workbooks, conversationally built dashboards, a creator mode for end users inside an embedded product — asks through it, and governance travels with it. Their phrasing for multi-tenancy is worth quoting because it is the design claim: \"Multi-tenant by construction. Governance flows from your model through to your customers' permissions.\" There is an open-source framework and a managed Cube Cloud, and MCP is first-class — the semantic layer is exposed to Claude, ChatGPT and Cursor, with a Slack agent and agent-to-agent capability. If your real problem is that five tools disagree about what revenue means, Cube is addressing exactly that and we are not.",[15,24,26],{"id":25},"the-comparison","The comparison",[20,28,29,30,34],{},"Every statement about Cube on this page describes what their own documentation, pricing page, licence file and engineering blog said ",[31,32,33],"strong",{},"as of 2026-10-09",", read directly rather than from a search summary. This category moves fast — check their site for the current state.",[36,37,38,53],"table",{},[39,40,41],"thead",{},[42,43,44,47,50],"tr",{},[45,46],"th",{},[45,48,49],{},"vibedasher",[45,51,52],{},"Cube",[54,55,56,73,84,136,167,203,214,238,249],"tbody",{},[42,57,58,62,70],{},[59,60,61],"td",{},"Who writes the dashboard",[59,63,64,65,69],{},"Your coding agent, as React components over SQL. We host that tree as a static, client-rendered bundle — React only, with none of your server code in the request path and no ",[66,67,68],"code",{},"npm install"," at build time, because the libraries come from a pinned list.",[59,71,72],{},"A person or Cube's own agents, over the semantic model: workbooks, dashboards, chat, or creator mode for embedded end users.",[42,74,75,78,81],{},[59,76,77],{},"Hosting and governance",[59,79,80],{},"A URL behind your login, or an embed inside your product. Per-viewer row filters are the embed lane and only the embed lane — enforced server-side from a signed token. A dashboard you publish publicly is the other case: its credential-free data lane serves one Parquet extract per dataset and snapshot, so every visitor gets every row of it.",[59,82,83],{},"Multi-tenant by construction, with governance flowing from the model to each customer's permissions. Embedding via chat APIs, iframes, creator mode and data APIs.",[42,85,86,89,100],{},[59,87,88],{},"MCP / agent control plane",[59,90,91,92,95,96,99],{},"An MCP server you install: ",[66,93,94],{},"pip install vibedasher-mcp"," then ",[66,97,98],{},"vibedasher login",". It covers the delivery loop — connect a source, push dashboard code, build, publish, schedule a refresh, eject. A refresh needs a dataset backed by a live connection, an uploaded file cannot be scheduled, and every cron is read in UTC.",[59,101,102,105,106,110,111,114,115,118,119,122,123,122,126,122,129,132,133],{},[31,103,104],{},"Stronger than ours on this row, and worth saying so."," They operate the endpoint rather than shipping you one: ",[107,108,109],"em",{},"\"Cube hosts an MCP server endpoint. MCP clients connect over HTTPS and authenticate via OAuth\"",", at ",[66,112,113],{},"https://cubecloud.dev/mcp",", and ",[107,116,117],{},"\"The MCP server is available on all plans\""," including the free tier. It also authors content — ",[66,120,121],{},"createWorkbook",", ",[66,124,125],{},"createReport",[66,127,128],{},"updateDashboard",[66,130,131],{},"publishDashboard"," — and they are explicit that it is not a privilege escalation: ",[107,134,135],{},"\"MCP is a new way to reach your data, not a new access surface.\"",[42,137,138,141,144],{},[59,139,140],{},"Eject-to-code",[59,142,143],{},"Yes. React source in your repo, still querying our API. A real ejected dashboard runs on this site.",[59,145,146,147,150,151,154,155,158,159,166],{},"Split, and the split is the thing to understand. The ",[31,148,149],{},"data model"," is genuinely portable, in their words: ",[107,152,153],{},"\"you can take a model built in the cloud product and run it in your own Cube Core deployment, and vice versa.\""," The ",[31,156,157],{},"dashboard"," is not: workbooks and reports are rows in Cube Cloud addressed by vendor-issued ids — ",[107,160,161,162,165],{},"\"A dashboard's chart widget points at a specific ",[66,163,164],{},"reportId","\""," — and the only dashboard export in their API reference renders PNG or PDF. So a comparison that calls both of us \"dashboards as code\" would be wrong.",[42,168,169,172,181],{},[59,170,171],{},"Self-host",[59,173,174,177,178],{},[66,175,176],{},"docker compose"," for one host, a Helm chart for Kubernetes, the same engine as the managed cloud. ",[31,179,180],{},"No production customer yet.",[59,182,183,184,187,188,191,192,195,196,199,200],{},"Real, under a permissive licence, and narrower than it sounds: ",[107,185,186],{},"\"Cube Core is the open-source semantic layer you can self-host with Docker\""," — the semantic layer and nothing else. Their CEO states the split in prose: the cloud product adds ",[107,189,190],{},"\"the frontend, the integrations, the deployment automation, the AI orchestration — instead of just the semantic layer.\""," There is ",[31,193,194],{},"no published feature matrix"," between the two; the nine Cube Core documentation pages mention MCP zero times, so read \"available on all plans\" as all ",[107,197,198],{},"Cloud"," plans. One named limit: open-source Cube Store ",[107,201,202],{},"\"doesn't support replicating any of its nodes.\"",[42,204,205,208,211],{},[59,206,207],{},"Managed cloud region",[59,209,210],{},"Data plane in the European Union, Frankfurt. Sign-in identity — email address and account membership — is held in the United States for every account; there is no region selector. The self-hosted build keeps both in your network.",[59,212,213],{},"Cube Cloud — check their docs for regions.",[42,215,216,219,235],{},[59,217,218],{},"Connectors",[59,220,221,222,227,228,234],{},"Warehouses, databases and SaaS sources, plus CSV and Excel upload. The count is on ",[223,224,226],"a",{"href":225},"/pricing","pricing","; the full list, including what is not available yet, is in the ",[223,229,233],{"href":230,"rel":231},"https://docs.vibedasher.com",[232],"nofollow","documentation",".",[59,236,237],{},"The warehouses and databases their data sources support — see their documentation.",[42,239,240,243,246],{},[59,241,242],{},"Licence",[59,244,245],{},"Proprietary, with a self-host option.",[59,247,248],{},"As of 2026-10-06, their licence file says: Apache License as the default throughout their repository, with some files under MIT per their headers; Cube Cloud is commercial.",[42,250,251,254,259],{},[59,252,253],{},"Pricing model",[59,255,256,257,234],{},"See ",[223,258,225],{"href":225},[59,260,261,262,122,265,268],{},"Published per seat, which ours is not: a free tier, ",[107,263,264],{},"\"Starter — $40 per developer / month\"",[107,266,267],{},"\"Premium — $80 per developer / month\"",", Enterprise on request. Embedded dashboards and an uptime commitment start at Premium; dedicated single-tenant and bring-your-own-cloud are Enterprise add-ons. Note their pricing page never mentions Cube Core or self-hosting — for that, their docs and licence file are the source.",[15,270,272],{"id":271},"choose-cube-if","Choose Cube if",[274,275,276,280,283,286],"ul",{},[277,278,279],"li",{},"The problem is metric disagreement across surfaces, and one governed model that everything asks through is the fix.",[277,281,282],{},"You want agents constrained to governed questions rather than given a SQL prompt and trusted.",[277,284,285],{},"You are building several analytics surfaces — internal, embedded, conversational — and want them to answer identically.",[277,287,288],{},"A semantic layer you can self-host under a permissive licence matters to your architecture.",[15,290,292],{"id":291},"choose-vibedasher-if","Choose vibedasher if",[274,294,295,298,301,309],{},[277,296,297],{},"The dashboard is React your agent already wrote, and you need it hosted as it is.",[277,299,300],{},"You want the agent to run the delivery loop end to end from a terminal, not only to ask questions.",[277,302,303,304,308],{},"The exit matters: ",[223,305,307],{"href":306},"/eject-demo","eject-to-code"," hands the dashboard back as source in your repo.",[277,310,311,312,234],{},"European residency or your own network is the constraint; see ",[223,313,315],{"href":314},"/self-hosted-bi","self-hosted BI",[20,317,318,321],{},[31,319,320],{},"And if both, which is common:"," point your agent's SQL at a governed layer and host the dashboard here. Nothing in either product forbids it.",[15,323,325],{"id":324},"next","Next",[20,327,328,333,334,338,339],{},[223,329,332],{"href":330,"rel":331},"https://cloud.vibedasher.com",[232],"Start building"," · ",[223,335,337],{"href":336},"/contact","Talk to us"," · back to ",[223,340,342],{"href":341},"/compare","compare",{"title":344,"searchDepth":345,"depth":345,"links":346},"",2,[347,348,349,350,351],{"id":17,"depth":345,"text":18},{"id":25,"depth":345,"text":26},{"id":271,"depth":345,"text":272},{"id":291,"depth":345,"text":292},{"id":324,"depth":345,"text":325},"Cube is an agentic analytics platform built on a semantic layer, governing what humans and AI agents may ask. vibedasher hosts the React and SQL your coding agent wrote. A sourced comparison — and why the two are often complementary rather than rival.","md",[355,358,361,364,367],{"q":356,"a":357},"Is Cube a competitor or something I run alongside?","More often the second, and we would rather say that than win a comparison on a false premise. Cube's job is the semantic layer: one governed model that every surface — a chart, an embedded product, an AI agent over MCP — has to ask through. Our job is hosting and governing the dashboard your agent wrote. A team that wants a single definition of a metric and a React dashboard an agent produced can reasonably want both.",{"q":359,"a":360},"Does vibedasher have a semantic layer?","No, and that is the honest gap in this comparison. A metric here is the SQL your agent wrote, reviewable in your repository, and nothing in the product stops two dashboards computing the same number two ways. If governed metric consistency across every surface is the problem you are solving, Cube is built for it and we are not.",{"q":362,"a":363},"Both of you talk about MCP. What is the difference?","Scope. Cube exposes the semantic model over MCP so an agent in Claude, ChatGPT or Cursor asks governed questions and gets governed answers. Ours exposes the delivery loop: connect a source, push dashboard code, build it, publish it, schedule a refresh, eject the source. One governs what may be asked; the other ships what was built.",{"q":365,"a":366},"What is each licensed under?","Checked on 2026-10-06, from the licence file in their public repository. Cube's public repository states Apache License as the default throughout, with some files under MIT where their headers say so, per its licence file; Cube Cloud is the commercial managed service. vibedasher is proprietary, with a self-host option. Check their docs for which capabilities live where.",{"q":368,"a":369},"Can Cube host the dashboard itself?","Cube's own positioning now includes dashboards, workbooks and a creator mode for end users inside an embedded product, so it does more than serve an API. What it does not do is host a React component tree your agent wrote — those dashboards are built in Cube, by a person or by its agents, over the semantic model.",{},true,"/compare/cube",{"title":5,"description":352},"vibedasher vs Cube — semantic layer and agent governance vs hosting your agent's React","compare/cube","3UPFoxELD99C6ize-lAOaabu6C2Ye_jzZ-Z28ofNpBk",{"id":378,"title":379,"author":6,"authorImage":6,"authorTitle":6,"body":380,"category":6,"date":6,"description":591,"extension":353,"faq":6,"image":6,"meta":592,"navigation":371,"path":341,"readTime":6,"seo":593,"seoTitle":594,"stem":342,"__hash__":595},"content/compare.md","Compare",{"type":8,"value":381,"toc":586},[382,386,403,406,416,420,431,489,492,498,520,523,550,560,562],[11,383,385],{"id":384},"vibedasher-compared","vibedasher compared",[20,387,388,389,392,393,392,396,399,400,234],{},"If you asked an answer engine for ",[107,390,391],{},"a Tableau alternative where my coding agent builds the dashboards",", or ",[107,394,395],{},"embedded analytics for a small SaaS where an AI agent does the building",[107,397,398],{},"self-hosted BI my agent can drive over MCP",", you got back some subset of six names: ",[31,401,402],{},"Metabase, Lightdash, Evidence, Apache Superset, Embeddable, Cube",[20,404,405],{},"They are all real products with real users, and several of them are better choices than we are for a reader who wants what they do. These pages say which reader that is, in each case, and where the actual difference lies.",[20,407,408,409,412,413,415],{},"The short version of the difference, before the detail. In every one of those products the dashboard is an object expressed in ",[31,410,411],{},"their"," model — a saved question, a chart definition, a semantic model, a framework-flavoured file. vibedasher hosts the artifact your coding agent already produced: React components over SQL, unmodified. That is the whole distinction, and everything else on these pages follows from it. One limit of ours belongs in that sentence: what we host is a static, client-rendered bundle — React only, none of your server code in the request path, no ",[66,414,68],{}," at build time.",[15,417,419],{"id":418},"the-six-pages","The six pages",[20,421,422,423,426,427,430],{},"One line each. Every factual claim about another product in them is as of ",[31,424,425],{},"2026-10-06","; the method is in ",[107,428,429],{},"How we compared",", below.",[274,432,433,442,451,460,469,478],{},[277,434,435,441],{},[31,436,437],{},[223,438,440],{"href":439},"/compare/metabase","vibedasher vs Metabase"," — the default answer to \"open-source BI\", with a visual query builder, an embedding SDK and an official MCP server. Choose them for self-service analytics across a whole company.",[277,443,444,450],{},[31,445,446],{},[223,447,449],{"href":448},"/compare/lightdash","vibedasher vs Lightdash"," — BI on top of your dbt project, metrics defined once in the semantic layer. Choose them if dbt is already the centre of your stack.",[277,452,453,459],{},[31,454,455],{},[223,456,458],{"href":457},"/compare/evidence","vibedasher vs Evidence"," — BI as code in a git repo: markdown files carrying SQL and components, push to deploy. Closest to us in philosophy, different in artifact.",[277,461,462,468],{},[31,463,464],{},[223,465,467],{"href":466},"/compare/superset","vibedasher vs Apache Superset"," — the Apache Software Foundation's exploration and visualisation platform, self-hosted by default, no vendor at all. Choose it if \"no vendor\" is the requirement.",[277,470,471,477],{},[31,472,473],{},[223,474,476],{"href":475},"/compare/embeddable","vibedasher vs Embeddable"," — customer-facing analytics for SaaS, dashboards as code, web components rather than iframes. The closest competitor on the embedded-analytics job.",[277,479,480,484,485,488],{},[31,481,482],{},[223,483,5],{"href":372}," — a semantic layer and agentic analytics platform that governs what every surface, human or agent, is allowed to ask. Often the right thing to run ",[107,486,487],{},"alongside"," a presentation layer rather than instead of one.",[15,490,429],{"id":491},"how-we-compared",[20,493,494,495,497],{},"Checked on ",[31,496,425],{},", from public sources only:",[274,499,500,507,513],{},[277,501,502,503,506],{},"each vendor's ",[31,504,505],{},"own homepage",", for how they position themselves — we describe them in their words, not in a straw-man of ours;",[277,508,502,509,512],{},[31,510,511],{},"public documentation",", for whether a capability exists (MCP server, embedding, row-level security, self-hosting) rather than whether it is advertised;",[277,514,515,516,519],{},"the ",[31,517,518],{},"licence file in their public repository",", where there is one, for the licence row of each table.",[20,521,522],{},"Three rules we held ourselves to:",[524,525,526,534,540],"ol",{},[277,527,528,531,532,234],{},[31,529,530],{},"No competitor prices."," Pricing models change weekly and a stale number in our favour is a lie with a date on it. Every table sends you to their own pricing page, and ours to ",[223,533,225],{"href":225},[277,535,536,539],{},[31,537,538],{},"No disparagement."," Each page opens with what the other tool is genuinely good at. If you read one and come away thinking \"they sound better for me\", the page worked.",[277,541,542,545,546,549],{},[31,543,544],{},"Where we are unsure, we say so."," A capability we could not confirm in their docs is written as ",[107,547,548],{},"check their docs",", not as an absence.",[20,551,552,553,556,557,559],{},"And the one about ourselves, which is the easiest to omit and the fastest to be caught in: the ",[31,554,555],{},"self-hosted path has no production customer yet",". It is documented and tested on ",[66,558,176],{}," and on a Helm chart, and nobody is running their business on it. That sentence is on every page below.",[15,561,325],{"id":324},[20,563,564,568,569,573,574,577,578,582,583,234],{},[223,565,567],{"href":566},"/embedded-analytics","Embedded analytics"," if the dashboards go in front of your own customers. ",[223,570,572],{"href":571},"/tableau-alternative","Tableau alternative"," if you are replacing something older. ",[223,575,576],{"href":314},"Self-hosted BI"," for the deployment shapes. To try it: ",[223,579,581],{"href":330,"rel":580},[232],"cloud.vibedasher.com",". To ask a person: ",[223,584,585],{"href":336},"contact",{"title":344,"searchDepth":345,"depth":345,"links":587},[588,589,590],{"id":418,"depth":345,"text":419},{"id":491,"depth":345,"text":429},{"id":324,"depth":345,"text":325},"Honest, sourced comparisons between vibedasher and the six tools most often named when someone asks for a BI or embedded-analytics platform a coding agent can drive: Metabase, Lightdash, Evidence, Apache Superset, Embeddable and Cube.",{},{"title":379,"description":591},"vibedasher compared: Metabase, Lightdash, Evidence, Superset, Embeddable, Cube","iefWyCOh3q9ilPovfO79RewFVL87D6Alsfg68yZwors",1791882564368]