No-code apps and databases decision
Bolt.new
A capable developer can build a narrow AI-to-static-app workflow (prompt → generated repo → deploy) in ~30 hours using OpenAI, Supabase and Vercel; reproducing Bolt’s enterprise hosting, token-routing agents, team features and integrated design-system at scale is much harder, so keeping Bolt may be warranted for full product needs.
Visit website↗Open-source builds that already do this
Every project below is open source and already does this job today. Fork one, self-host it, or take the parts you need — the build prompt further down assumes an empty file, and this is the shortcut past that. Licences differ; check the one on each card before you ship.
What a replacement has to do
- User prompts → AI generates project files (frontend + backend stubs) → deploy to hosting; plus basic auth and database wiring.
What it still won’t have
- Enterprise-grade Bolt Cloud hosting, analytics and built-in infra
- Bolt’s claimed token-routing and agent orchestration
- Integrated design-system & visual Plan mode
- Team management, SSO, audit logs and enterprise support
- Unlimited built-in databases and UI for DB management
What remains hard
- Brand trust
Trusted by Google, Microsoft, Salesforce, AWS, Meta, HubSpot, Accenture, TikTok, Shopify, Chegg, Cloudflare, Netlify, FactSet, ByteDance, PwC, Mozilla, Intel, Xano, Stripe
- Infrastructure at scale
Bolt Cloud gives you enterprise-grade backend infrastructure including hosting, databases, integrations and more.
First-year cost
Keep paying
Paying is—cheaper in year one.
On cash alone, building overtakes the subscription at 7 seats.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a minimal AI-first website/app builder using Next.js (TypeScript), React, Tailwind CSS, Node backend (Express) and a Supabase database. Core features in scope: 1) simple chat UI that accepts a project prompt and basic settings; 2) call OpenAI to generate project files (React app skeleton + package.json + simple API route); 3) store generated files in a Git repo (use GitHub API) or object store; 4) trigger automated build & deploy to Vercel/Netlify via their API and return a preview URL; 5) wire a Supabase DB and simple email/password auth scaffold into generated app; 6) show file browser and logs, and run a lint/test step before deploy. Out of scope: multi-user billing/teams, enterprise SSO/audit logs, custom agent orchestration, visual Plan mode, image-editing UIs. Include error handling for LLM failures, rate limits, and deployment errors; include automated tests for generation → build → deploy flow and README with setup steps and env var list.
How we checked
How the score was reached
- Partly verdict base52
- An open-source build was found+5
- 5 cited sources+3
- Price verified on pricing page+3
- 3/3 assessment runs agreed+4
- Hard moats found in the evidence-3
- Evidence score64
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time — so the same evidence always produces the same number.
How scoring works →Cited sources · 5
Every page the run actually retrieved.
- official productBolt AI builder: Websites, apps & prototypes
- official pricingPlans & pricing: Bolt’s AI powered website and app builder
- official docsBolt Help Center - Bolt
- open sourceBudibase/budibase
- open sourcenocobase/nocobase
Integrity checks
What held up, and what did not.
