Earlier this year, Apple introduced its Foundation Model Framework during WWDC 2025, allowing developers to use the company’s local AI model that is for power features in its applications.
The company said that with this structure, developers gain access to the AI model without worrying about any estimate cost. In addition, these local models have capabilities like directed generation and equipment calling.
As the iOS 26 is rolling to all users, developers are updating their apps to include the features operated by the local AI model of Apple. Apple’s models are smaller than the leading models of Openai, Anthropic, Google or Meta. This is the reason that local-care facilities improve the quality of life to a great extent with these apps rather than presenting major changes in the workflow of the app.
Below are some of the first apps to tap in Apple’s AI framework.
Lill artist
Lill artist The app offers children to help children learn various skills such as creativity, mathematics and music. Developers Arima Jain and Aman Jain sent the AI story manufacturer with iOS 26 update. This allows users to select a character and a subject, in which the app generates a story using AI. The developer said that the text generation in the story is run by the local model.
Day
Developer of Day The app is working on a prototype to suggest emojis automatically for timeline events based on the title for the Daily Planner app.
Moneycuch
Finance tracking app Moneycuch There are two clean -cotton features operated by local models. First, the app shows insight about your expenses, such as have you spent more than average on grocery items for that particular week. Other features automatically suggest categories and subcontinions for a spending items for quick entries.
Look up
Word learning app Look up There are two new modes using Apple’s AI model. There is a new learning mode, which takes advantage of a local model to create an example of a word. Also, the example asks users to explain the use of the word in a sentence.
The developer is also using an on-device model to generate a map of the origin of a word.
Work
Like some other apps, Work The app automatically implemented a feature to suggest tag for an entry using local models. It is using these models to detect a recurring function and schedule it accordingly. And the app allows users to speak some things and uses local models to break them in various tasks without using the Internet.
one day
Automatic-owned jernling app one day Using the model of Apple to get highlight and is suggesting the title for your entry. The team has also implemented a feature to generate the signals that you have already written to write more depending on what you have already written.
Crauton
Recipe app Crauton Apple is using intelligence to suggest tags for a recipe and specify the timer’s name. It uses AI to break a block of the text at an easy-to-bey stages of cooking.
Sinusise
Digital signature app Sinusise The major insights from a contract are using the local model of Apple to extract and summarize users the document they are signing.
Dark noise
Background sound app Dark noise Users have used the local model of Apple to describe a soundscape in a few words and generate one based on it. Once made, you can adjust the levels of different elements of the soundscape.
lights out
lights out There is a new app to track F1 season and Grand Prix from Twitter Client Every and Mastodon Client Mammath developer Shihab Mehboob. The app uses an on-device AI model to summarize the comment during a race.
Capture
Note -taking app Capture Users are using local AIs to show the category suggestions as they type in their notes or functions.
Lumi
Sun and weather-tracking app Lumi Now using AI shows the weather related suggestions on the app.
Cardpoiners
Cardpoiners There is an app that helps you track the expenses of your credit card and suggests the best way to earn points from you. The new version of the app is using AI to ask users questions about their cards and offers.
Guitar
Guitar In some ways the Apple Foundation is a guitar learning app using model framework. Users will see an explanation of a raga when learning it. The app is showing the insight of advanced players based on time intervals. Also, the AI model is helping the developer support in 15 languages.
SmartGime App The local AI is used to convert the details of the workout to a step-wise set with representative count, intervals and equipment. It also gives users a workout summary with monthly progression, regular breakdown and personal exercise performance.
Moody
Jernling app Moody Using the model of Apple to give users a personal signal based on its mood logging. Models can also help users briefly present the post, find and organize previous entries and organize them.
Swing
This app helps the players of racket sports such as tennis and pickles improve their form based on video recording. App manufacturers are now using Basic model to give actionable and specific feedback,
Joho
India-based productivity suit company Zoho is using local model like its apps for power summary, translation and transcription Notebook for documents and tables for spreadsheets,
Trainfit
Workout app When you do not have special equipment available, you are using an on-device model to suggest an option for an exercise.
Material
Is two-do app Listen a listen mode It uses the AI model of Apple to hear the user and convert voice into personal tasks.
We will continue to update this list as we search for more apps using Apple’s local model.
