Photograph a menu, extract every item and price with your own AI key, review, then export POS-ready data.
Every restaurant onboarding, POS migration, delivery-platform signup, and print-menu update starts the same way: someone retypes the entire menu. Item by item, price by price, section by section. A 120-item menu with modifiers is an afternoon of data entry — and it happens again at every price change.
Metered document-AI APIs can read menus, but they charge per page forever, and your menu data flows through a third party's pipeline.
Extract sections, item names, descriptions, prices, modifiers, allergens, and dietary tags from a photo, scan, or PDF.
Choose OpenAI, Anthropic, or local Ollama and switch providers any time in Settings.
Edit any item, fix any price, and re-order or re-section everything before export.
Export nested CSV, flat POS-import CSV with auto-generated SKUs, or JSON.
Keep breakfast, lunch, dinner, and seasonal versions, and re-extract whenever prices change.
Runs on PHP 8+ with MySQL or SQLite on standard shared hosting, with no subscription or per-page metering beyond your own LLM cost.
Checkout isn't open yet. Leave your email and we'll send one note the moment Menura can be purchased — nothing else.
Two-minute web installer; PHP 8+ with MySQL or SQLite, standard shared-hosting compatible.
Yes — Menura requires an API key from OpenAI, Anthropic, or a local Ollama instance. Your key is stored in your own database and transmitted only in the outbound request you trigger; it's never logged, echoed back to the browser, or passed through any Menura infrastructure.
Yes. Menura requires PHP 8+ with pdo, curl, and fileinfo — extensions present on virtually all cPanel/Plesk hosts. Installation takes about two minutes through the web installer.
No — Menura exports CSV/JSON, which you import manually into your POS.
No — Menura is human-in-the-loop by design; you review and correct everything on screen before export.
Handwritten menus in poor lighting may extract poorly. PDFs are supported natively by Anthropic; OpenAI and Ollama treat PDFs as best-effort, so for best PDF results use Anthropic or convert pages to PNG first.
The live demo is being prepared — check back shortly.
Demo — coming soon