Layla vs TREK

Layla is an AI travel agent and trip planner for building personalized itineraries across flights, hotels, activities, dining, and recommendations. It says it compares real-time prices and availability for flights, hotels, trains, and car rentals. Travelers can revise routes, schedules, and costs, then download PDF itineraries for offline access or sharing. Planning tools include interactive video, maps for multi-destination routes, a road trip planner, and creator travel videos connected to planning. Human travel advisors can plan, book, and manage trips; advisor help is described as available at no extra planning cost. Layla offers booking links and collaborative planning, and names Booking.com, Skyscanner, and GetYourGuide as travel partners. With explicit consent, users can import trip details from Gmail or Outlook and travel-profile information from Instagram. It runs on web, iOS, and Android, and has a free plan with no subscription fee; users pay for travel they choose to book. Layla cautions that AI-generated travel details may be incomplete, outdated, or inaccurate, so important information should be verified.

TREK is a self-hosted travel planner for organizing trips collaboratively in real time. Users can arrange places by day, optimize routes by distance, and send a day to Google Maps for turn-by-turn directions. Collaborators can edit itineraries together, see live presence, chat, run polls, track attendance, and share read-only public links. Reservation records hold confirmation numbers, files, and statuses for flights, stays, and restaurants. Budgets support multiple currencies and per-person or per-day totals; packing lists include templates and item assignments. OpenStreetMap is the default place search, Google Places is optional, and 16-day forecasts use Open-Meteo without API keys. Optional addons include travel statistics, vacation planning, journaling, and MCP integration. TREK runs on operator-controlled hardware, stores trip data in a SQLite file, and supports scheduled snapshots with retention policies. It can be installed as a PWA and has deployment documentation for Docker, Kubernetes, Unraid, Proxmox, Raspberry Pi, and Apple Silicon, with AMD64 and ARM64 images. It is open source under AGPL-3.0 and free for personal and internal company use.