robots.txt
# As a condition of accessing this website, you agree to abide by the
# following content signals:
#
# (a) If a Content-Signal = yes, you may collect content for the
# corresponding use.
#
# search = building a search index and linking to this site
# ai-input = using this content to ground or answer an AI query
# ai-train = training or fine-tuning AI models
# Everyone not named below.
User-agent: *
Content-Signal: search=yes,ai-input=yes,ai-train=yes
Allow: /
Disallow: /admin
Disallow: /auth/
Disallow: /cdn-cgi/
# Search engines.
User-agent: Googlebot
User-agent: Googlebot-Image
User-agent: Googlebot-Video
User-agent: Googlebot-News
User-agent: Storebot-Google
User-agent: Google-InspectionTool
User-agent: AdsBot-Google
User-agent: Mediapartners-Google
User-agent: bingbot
User-agent: msnbot
User-agent: BingPreview
User-agent: AdIdxBot
User-agent: Slurp
User-agent: DuckDuckBot
User-agent: Baiduspider
User-agent: YandexBot
User-agent: YandexImages
User-agent: Yeti
User-agent: SeznamBot
User-agent: Sogou
User-agent: Qwantbot
User-agent: MojeekBot
User-agent: Coccocbot
User-agent: archive.org_bot
User-agent: ia_archiver
Content-Signal: search=yes,ai-input=yes,ai-train=yes
Allow: /
Disallow: /admin
Disallow: /auth/
Disallow: /cdn-cgi/
# Link preview fetchers.
User-agent: Twitterbot
User-agent: LinkedInBot
User-agent: Slackbot
User-agent: Slack-ImgProxy
User-agent: Discordbot
User-agent: WhatsApp
User-agent: TelegramBot
User-agent: Pinterestbot
User-agent: redditbot
User-agent: Embedly
User-agent: Iframely
User-agent: SkypeUriPreview
User-agent: vkShare
Content-Signal: search=yes,ai-input=yes,ai-train=yes
Allow: /
Disallow: /admin
Disallow: /auth/
Disallow: /cdn-cgi/
# AI crawlers, assistants and agents. All welcome: being quoted by an assistant is a distribution channel, not a leak.
User-agent: AddSearchBot
User-agent: AgentDataBot
User-agent: AgentTimes
User-agent: AI2Bot
User-agent: AI2Bot-DeepResearchEval
User-agent: Ai2Bot-Dolma
User-agent: aiHitBot
User-agent: AIWebIndex
User-agent: amazon-kendra
User-agent: amazon-QBusiness
User-agent: Amazonbot
User-agent: AmazonBuyForMe
User-agent: Amzn-SearchBot
User-agent: Amzn-User
User-agent: Andibot
User-agent: Anomura
User-agent: anthropic-ai
User-agent: ApifyBot
User-agent: ApifyWebsiteContentCrawler
User-agent: Applebot
User-agent: Applebot-Extended
User-agent: Aranet-SearchBot
User-agent: atlassian-bot
User-agent: Awario
User-agent: AzureAI-SearchBot
User-agent: bedrockbot
User-agent: bigsur.ai
User-agent: BixelBot
User-agent: Bravebot
User-agent: Brightbot
User-agent: Brightbot 1.0
User-agent: BuddyBot
User-agent: Bytespider
User-agent: CCBot
User-agent: Channel3Bot
User-agent: ChatGLM-Spider
User-agent: ChatGPT Agent
User-agent: ChatGPT-User
User-agent: Claude-Code
User-agent: Claude-SearchBot
User-agent: Claude-User
User-agent: Claude-Web
User-agent: ClaudeBot
User-agent: Cloudflare-AutoRAG
User-agent: CloudflareBrowserRenderingCrawler
User-agent: CloudVertexBot
User-agent: Code
User-agent: cohere-ai
User-agent: cohere-training-data-crawler
User-agent: Cotoyogi
User-agent: CragCrawler
User-agent: Crawl4AI
User-agent: Crawlspace
User-agent: Cursor
User-agent: Datenbank Crawler
User-agent: DeepSeekBot
User-agent: Devin
User-agent: Diffbot
User-agent: Diffbot-User
User-agent: DoubaoBot
User-agent: DuckAssistBot
User-agent: Echobot Bot
User-agent: EchoboxBot
User-agent: ERNIEBot
User-agent: ExaBot
User-agent: ExaSearchBot
User-agent: FacebookBot
User-agent: facebookexternalhit
User-agent: Factset_spyderbot
User-agent: FirecrawlAgent
User-agent: FriendlyCrawler
User-agent: GeistHaus-PageFetcher
User-agent: Gemini-Deep-Research
User-agent: Google-Agent
User-agent: Google-CloudVertexBot
User-agent: Google-Extended
User-agent: Google-Firebase
User-agent: Google-Gemini-CLI
User-agent: Google-NotebookLM
User-agent: GoogleAgent-Mariner
User-agent: GoogleAgent-URLContext
User-agent: GoogleOther
User-agent: GoogleOther-Image
User-agent: GoogleOther-Video
User-agent: GPTBot
User-agent: HenkBot
User-agent: iAskBot
User-agent: iaskspider
User-agent: iaskspider/2.0
User-agent: ICC-Crawler
User-agent: ImagesiftBot
User-agent: imageSpider
User-agent: img2dataset
User-agent: ISSCyberRiskCrawler
User-agent: kagi-fetcher
User-agent: Kangaroo Bot
User-agent: KeenableBot
User-agent: Kimi-Agent
User-agent: Kimi-SearchBot
User-agent: Kimi-User
User-agent: KimiBot
User-agent: KlaviyoAIBot
User-agent: KunatoCrawler
User-agent: laion-huggingface-processor
User-agent: LAIONDownloader
User-agent: LCC
User-agent: Lightpanda
User-agent: LinerBot
User-agent: Linguee Bot
User-agent: LinkupBot
User-agent: Manus-User
User-agent: meta-externalagent
User-agent: meta-externalfetcher
User-agent: meta-webindexer
User-agent: MistralAI-Index
User-agent: MistralAI-Training
User-agent: MistralAI-User
User-agent: MistralAI-User/1.0
User-agent: Mozilla-Tabstack
User-agent: MyCentralAIScraperBot
User-agent: NagetBot
User-agent: netEstate Imprint Crawler
User-agent: newsai
User-agent: NotebookLM
User-agent: NovaAct
User-agent: OAI-AdsBot
User-agent: OAI-SearchBot
User-agent: omgili
User-agent: omgilibot
User-agent: OpenAI
User-agent: opencode
User-agent: Operator
User-agent: PanguBot
User-agent: Panscient
User-agent: panscient.com
User-agent: Perplexity-User
User-agent: PerplexityBot
User-agent: PetalBot
User-agent: PhindBot
User-agent: Poggio-Citations
User-agent: Poseidon Research Crawler
User-agent: qodercli
User-agent: QualifiedBot
User-agent: Querit-SearchBot
User-agent: QueritBot
User-agent: QuillBot
User-agent: quillbot.com
User-agent: QwenBot
User-agent: Reflectionbot
User-agent: SBIntuitionsBot
User-agent: Scrapy
User-agent: SemrushBot-OCOB
User-agent: SemrushBot-SWA
User-agent: Shap-User
User-agent: ShapBot
User-agent: Sidetrade indexer bot
User-agent: Spider
User-agent: TavilyBot
User-agent: Terra Cotta
User-agent: TerraCotta
User-agent: Thinkbot
User-agent: TikTokSpider
User-agent: Timpibot
User-agent: TongyiBot
User-agent: Trae
User-agent: TwinAgent
User-agent: UseAI
User-agent: VelenPublicWebCrawler
User-agent: WARDBot
User-agent: Webzio-Extended
User-agent: wpbot
User-agent: WRTNBot
User-agent: YaK
User-agent: YandexAdditional
User-agent: YandexAdditionalBot
User-agent: YiyanBot
User-agent: YouBot
User-agent: ZanistaBot
Content-Signal: search=yes,ai-input=yes,ai-train=yes
Allow: /
Disallow: /admin
Disallow: /auth/
Disallow: /cdn-cgi/
Sitemap: https://orlume.site/sitemap.xml
llms.txt
# Orlume
> Orlume is a design-first AI website builder. You describe a site in plain language; Orlume makes the design decisions first (layout, typography, hierarchy and spacing) and then builds a production-ready React site from them. Each project runs as a working preview until you publish it to its own orlume.site address.
Most AI website builders generate code and treat design as a side effect, which is why their output converges on the same centered hero, three feature cards and pricing table. Orlume runs a separate design pass before any code is written, so its sites look designed rather than generated. It is aimed at founders, indie makers and small teams who want a site that does not look AI-made and do not have a designer.
## What it does
- Builds landing pages, portfolios, SaaS marketing sites, dashboards, storefronts and web apps.
- A site starts from a text prompt, or on paid plans from an uploaded screenshot or design.
- Think Mode spends extra reasoning on design decisions before writing any code, covering hierarchy, brand personality and type contrast, for pages that need to be right rather than fast.
- Dream Mode builds a bento grid with real hierarchy and an embedded product video in one click.
- Both modes are switches on the prompt box and are available on paid plans.
- Changes are made by asking in plain language. Click Publish in the workspace, pick a name, and the site is live at name.orlume.site.
- The Free plan includes 10 message credits and builds with the standard engine. Paid plans start at $10 a month. They come with more credits, generate with the full design engine and add Think Mode, Dream Mode and image uploads.
- Code download starts on the Personal plan.
- Custom domains are part of the Pro plan.
## Pricing (USD)
- Free ($0): 10 message credits, Unlimited sites. Not included: Full design engine, Think mode, Dream mode, Image uploads.
- Starter ($10 per month): 400 message credits, Full design engine, Think mode, Dream mode, Image uploads, Unlimited sites.
- Personal ($25 per month): 1000 message credits, Everything in Starter, Credit rollovers, In-app code edits, Code download.
- Pro ($50 per month): 2500 message credits, Everything in Personal, GitHub integration (coming soon), Priority support, Custom domains. To upgrade to this plan, email contact@orlume.site.
## Pages
- [Home](https://orlume.site): the builder itself, the feature tour and the FAQ.
- [Pricing](https://orlume.site/pricing): the plans above in full.
- [Blog](https://orlume.site/blog): articles on AI web design and comparisons with other AI website builders.
- [Privacy policy](https://orlume.site/privacy)
- [Terms and conditions](https://orlume.site/terms)
## Articles
- [Replit vs Base44: I’ve Used Both & Here’s My Take](https://orlume.site/blog/replit-vs-base44-my-take): Replit and Base44 both turn prompts into apps, but they favor different workflows: code-level control and extensibility versus a lower-friction path from idea to interactive prototype. Compare their trade-offs, evaluate them with the same real-world test, and verify security, ownership, and deployment requirements before shipping.
- [v0 vs Cursor: I’ve Used Both So You Don’t Have To!](https://orlume.site/blog/v0-vs-cursor-practical-workflow-comparison): v0 is strongest when you need to turn a product idea into a visible interface; Cursor is the more natural fit for making changes inside an existing codebase. This practical comparison breaks down their workflows, trade-offs, review risks, and how to combine them without confusing a convincing prototype for production-ready software.
- [v0 vs Cursor: I’ve Used Both So You Don’t Have To!](https://orlume.site/blog/v0-vs-cursor-workflow-comparison): v0 and Cursor can both generate code, but they start from different contexts: v0 excels at exploring interfaces, while Cursor is designed for repository-aware implementation. This practical comparison explains when to use each, how to combine them, and how to evaluate their output with review and tests.
- [How to Make Beautiful Websites with AI: A Design-to-Code Workflow That Holds Up](https://orlume.site/blog/make-beautiful-websites-with-ai): Learn a practical AI-assisted workflow for turning a clear product brief into a coherent, responsive website. Build a small design system, generate changes in reviewable increments, and validate the rendered result for accessibility, performance, and correctness before launch.
- [I Generated 50 Websites With AI. They Were All the Same Website.](https://orlume.site/blog/lovable-alternatives): Looking for a Lovable alternative? Why every AI website builder ships the same purple gradient and three feature cards — and what it takes to actually design.
## Frequently asked questions
### What is Orlume?
Orlume is an AI website builder that designs before it writes code. You describe what you want, and it makes real design decisions about hierarchy, type scale, spacing and layout, then builds a production-ready React site from them. No coding or design skills needed.
### How is Orlume different from other AI website builders?
Most AI builders generate code and treat design as a side effect, which is why their output converges on the same centered hero, three feature cards and pricing table. Orlume runs a separate design pass first, so the layout is decided for your brief. You can test any builder the same way: run one prompt twice and compare the two layouts.
### Is Orlume free to use?
Yes, you can build and publish without paying. The Free plan includes 10 message credits and builds with the standard engine. Paid plans start at $10 a month. They come with more credits, generate with the full design engine and add Think Mode, Dream Mode and image uploads. You can compare them on the pricing page.
### What are Think Mode and Dream Mode?
Think Mode spends extra reasoning on design decisions before writing any code, covering hierarchy, brand personality and type contrast, for pages that need to be right rather than fast. Dream Mode builds a bento grid with real hierarchy and an embedded product video in one click. Both modes are switches on the prompt box and are available on paid plans.
### Do I need to know how to code?
No. You describe the site in plain language and ask for changes the same way. If you do want the code, it is a full React project that you own. Code download starts on the Personal plan.
### What kinds of websites can I build?
Landing pages, portfolios, SaaS marketing sites, dashboards, e-commerce storefronts, and web apps. Orlume is strongest on anything judged visually, where design quality is the point.
### How do I get my site online?
Click Publish in the workspace, pick a name, and the site is live at name.orlume.site. Until then each project runs as a working preview that you can check at desktop, tablet and phone widths. There is no hosting to set up.
### Can I use my own domain?
Publishing gives every site its own orlume.site address. Custom domains are part of the Pro plan. You can also download the code, from the Personal plan up, and host it anywhere you like.
## Contact
- Email: contact@orlume.site
- X: https://x.com/orlume
- LinkedIn: https://www.linkedin.com/company/orlume/
- Product Hunt: https://www.producthunt.com/products/orlume
