PagentRequest a testRequest

AI assistants for meal planning and grocery lists

Which AI assistants can help with meal planning and grocery lists?

Updated · 4 assistants ranked

Four assistants in our catalog mention meals, groceries or lists: Ollie (email, calendar, reminders, meals), Orbits (family calendars, chores, meals), OpenInstinct (ordering groceries) and Nessi (reminders and lists). All four took at least two of the five tests that matter here, and are ranked. We do not test recipes, nutrition or diets.

Top picks

  1. 1
    OllieTop pickFastest that's still good

    Tested on 4 of 5Best: Routines9, Online tasks9Time to finish 7m 25s

    Free tier · iMessage, SMS · 76% positive in 44 public quotes

    8.0
  2. 2
    OpenInstinctBest free option

    Tested on 3 of 5Best: Online tasks10, Connected apps6

    Free · iMessage, App

    6.7
  3. 3

    Tested on 3 of 5Best: Online tasks7, Memory6

    Free · iMessage, App

    5.7
  4. 4

    Tested on 2 of 5Best: Connected apps5, Memory4

    Free · iMessage, SMS, App

    4.5

Top pick: highest average here. Best free option: highest among assistants you can use for free. Fastest that's still good: quickest median time to finish among those averaging 7 or more.

Full ranking

#AssistantRoutinesMemoryPurchasingOnline tasksConnected appsAverage
1Ollie9—7978.0
2OpenInstinct—4—1066.7
3Orbits—647—5.7
4Nessi—4——54.5

How this report works

Included are assistants whose catalog description (checked 2026-10-26) mentions meals, groceries or lists. They are ranked by the average of five tests: Routines (a weekly plan on time), Memory (your diet and dislikes), Purchasing (the grocery order), Online task (a job on a real website) and Connected apps (putting a list in the right place), with at least two taken. Scores come from the published score sheet and are dated 2026-10-27. 1 of the 4 ranked assistants (Nessi) carry pre-filled scores, which are placeholder estimates the site has not yet replaced with its own test runs, so treat their places as provisional.

How to choose

  • Decide whether you want a plan or an order. Ollie and Orbits describe meals; OpenInstinct describes ordering groceries, so check which half you need.
  • Check Purchasing before letting it pay: it must find the right item, use the right address and card, and confirm before paying.
  • Watch the price and limits. Ollie lists 50 messages a month plus 6 free a day, Orbits is free for people who join now, and OpenInstinct is free software where you pay for the services it runs on.

Can an AI assistant plan my meals for the week?

Two catalog entries say so: Ollie (meals, with email, calendar and reminders) and Orbits (family calendars, chores and meals). We do not test recipes or nutrition; we test whether the assistant runs a weekly job on time and remembers your preferences.

Can it order my groceries?

OpenInstinct describes ordering groceries and is open source, free and MIT licensed. Ollie and Orbits have Purchasing scores too (7 and 4), which measure finding the right item and confirming before paying, not a specific store.

Which of the four has the strongest scores?

Ollie averages 8.0 over four of the five tests, OpenInstinct 6.7 and Orbits 5.7, none of them pre-filled. Nessi (4.5, two tests) is pre-filled, so its place is provisional.

What users say

OpenInstinct · @DennisonBertram ·

I just setup and it ordered me lunch, running on my own infra.

Open source
Ollie · usecarly ·

Use Ollie if you want a text-only assistant that quietly watches your inbox and calendar and pings you; use Ohai if you want an app-based shared household workspace the whole family uses together.

Open source
Ollie · @EricaDiamond ·

The Text I Didn’t Know I Needed Every Morning

Open source

Frequently asked questions

Which assistant is best for this?

Ollie leads with an average of 8.0, ahead of OpenInstinct and Orbits.

Is there a good free option?

Ollie is the highest-ranked assistant with a free plan, averaging 8.0.

How did you test these assistants?

Every assistant got the same published tests, word for word, scored from three checkpoints and the number of corrections it needed. Each test is on the Dimensions page.

Related reports