Analytics cookies only (Google Analytics, PostHog). No ad networks, no cross-site tracking. See our privacy policy.
upstream: github.com/TomBursch/kitchenowl (opens in new tab) · github.com/mealie-recipes/mealie (opens in new tab)
Pick KitchenOwl
If the shared shopping list is the main job: real-time sync, native apps, expense balances and a simpler setup, with the smaller disk figure we measured (~1.4 GB against ~1.7 GB), accepting alpha-stage software.
Pick Mealie
If recipes come first: URL import, cookbooks, planner rules, and lower idle RAM (~202 MB against ~403 MB).
KitchenOwl and Mealie both manage recipes and meal plans for a household, but they start from different places. Mealie is a recipe manager first, built around importing recipes from the web and organising them. KitchenOwl is a grocery list first, with synced shopping lists, native mobile and desktop apps and household expense tracking, and recipes feed the list. KitchenOwl describes itself as a public alpha. If you are weighing Mealie against another recipe manager, Mealie vs Tandoor covers that.
KitchenOwl's README: "KitchenOwl is a smart self-hosted grocery list and recipe manager." Its feature list includes "Add items to your shopping list and sync them in real-time with multiple users", "Partial offline support", "Create a meal plan to always know what you'll be eating" and "Manage balances and track expenses of your household". It adds: "Please keep in mind that this project is still in development." The status list marks Public Alpha as current and Public Beta as not reached.
Mealie's README: "Mealie is a self hosted recipe manager, meal planner and shopping list with a RestAPI backend and a reactive frontend built in Vue for a pleasant user experience for the whole family." Its key features include recipe imports from a URL or by hand, a meal planner, a shopping list organised into supermarket sections, and cookbooks.
Mealie's docs go further on the recipe side. They describe a scraper that creates recipes from hundreds of websites from a URL, import from HTML or JSON, a manual editor, optional AI import, and migrations from sources such as Tandoor, Nextcloud Cookbooks and Paprika. Recipes can be organised with categories, tags, tools and cookbooks, and the meal planner has rules that pick random recipes by tag or category.
KitchenOwl's website says it stores recipes from the web or created by you, learns the order you usually tick off items, and suggests recipes. Its distinguishing parts are household sync and balances: after a shop you enter the expense and divide it between members.
Mealie's install docs say "SQLite is sufficient for most use cases" and call it ideal for 1 to 20 users; PostgreSQL "might be considered if you need to support many concurrent users", and some features, "such as fuzzy search", are only enabled on PostgreSQL. Our snippet is the SQLite one, a single container.
KitchenOwl's catalog entry lists a Flask backend on SQLite or PostgreSQL behind a Flutter web front end. Our snippet runs two services, a front end and a backend, with SQLite in a named volume.
Both are AGPL-3.0. If you modify either and offer it to others over a network, the AGPL requires you to share your changes.
We list 256 MB of RAM for KitchenOwl and 1 GB for Mealie; neither project documents
a minimum, so both are our estimates. Mealie's docs do recommend an explicit memory
limit, and the compose file sets 1000M, because Python can otherwise idle at higher
memory use. In our install check (GCP e2-standard-2, Ubuntu 26.04), idle at first
boot with the default install snippet and no real workload, KitchenOwl used
~403 MB of RAM and ~1.4 GB of disk (November 2026), and Mealie used
~202 MB of RAM and ~1.7 GB of disk (September 2026). The readings are from
different months, and disk is docker system df for the install.
KitchenOwl is 1 out of 5: a two-service compose file, a generated
JWT_SECRET_KEY, and the first visit creates the admin account. Mealie is 2 out
of 5: one container on port 9925, a BASE_URL to set, and a documented default
login (changeme@example.com) to change at once.
Choose KitchenOwl if the household shopping list is the main job: real-time sync, native apps, expense balances, a meal plan, the smaller disk we measured and a lower difficulty rating, accepting alpha-stage software.
Choose Mealie if recipes are the main job: URL import, cookbooks, planner rules, and the lower idle RAM we measured.
Both installs are verified on Ubuntu 26.04; the snippets are on the
KitchenOwl and Mealie pages. Back up
the KitchenOwl kitchenowl_data volume or Mealie's mealie-data volume (Mealie also
has backups in its UI), and put TLS in front of either.
KitchenOwl lists a lower minimum: 256 MB vs 1024 MB for Mealie.
KitchenOwl when the household grocery list, synced between members with expense tracking, is the main job (~403 MB RAM idle in our first-boot check). Mealie when recipe import and organisation come first (~202 MB RAM idle). KitchenOwl describes itself as a public alpha.
Its README says the project is still in development, and its status list ticks Public Alpha while Public Beta, described as stable with most planned features complete, is unticked.
Yes. Mealie's docs describe a recipe scraper that creates recipes from hundreds of websites given a URL, plus HTML or JSON import, a manual editor and optional AI import.
Mealie's docs say SQLite is sufficient for most use cases and ideal for 1 to 20 users. PostgreSQL suits many concurrent users, and some features such as fuzzy search are only enabled there.
Paid link — we earn a commission if you shop through it.
The single-container recipe app vs. the Postgres-backed power-user one.
A whiteboard-and-docs workspace on a plain Compose file vs. native apps on a commercial server.
A full identity platform vs. a passkey-only sign-in box.