On-Device AI Integration for iOS Apps

On-device AI is no longer a lab demo. Since iOS 18.4, Apple Intelligence is available in English. With iOS 26 and the Foundation Models framework, developers have access to a ~3 billion parameter language model running directly on the device — no API keys, no per-request costs, no internet connection required.

At AtalayaSoft we build intelligent features inside iOS apps: automatic summaries, content classification, conversational assistants, entity extraction and natural language-guided flows. With attention to what matters in production: native UX, reliability, user privacy and cost optimisation.

Gartner predicts 40% of enterprise apps will feature AI agents by end of 2026 — up from under 5% in 2025. Adoption is accelerating. The question is not whether your app needs intelligent features, but when you implement them.

IA en dispositivo para iOS

What we can build in your app

Features with Apple Foundation Models
Implementation of features using Apple's on-device model (~3B parameters): summaries, entity extraction, structured text generation, classification and tool calling. All with @Generable and @Guide macros for typed and predictable output.
In-app conversational assistants
Design and implementation of conversational flows integrated into the app: support chatbots, navigation assistants, natural language search. With context management, fallbacks and native UX — not a WebView with a generic chatbot.
Cloud LLM integration (Claude, GPT)
When the use case requires capabilities beyond the on-device model, we integrate LLM APIs such as Claude or GPT with robust error handling, intelligent caching, response streaming and per-request cost optimisation.
Core ML and custom models
Integration of custom machine learning models via Core ML: image classification, object detection, sentiment analysis, recommendation models. Conversion from PyTorch/TensorFlow and optimisation for Apple Silicon.
Privacy and local processing
Design of architectures that maximise on-device processing to comply with GDPR and user privacy expectations. Sensitive data never leaves the iPhone — AI runs where the data is.
API cost optimisation
Strategies to reduce AI API costs: response caching, batch processing, on-device models for simple tasks and cloud APIs only for complex tasks. Hybrid architecture balancing capability and cost.
iOS Development
AI Integration
Apple Development

The AI stack we use in iOS

We work with the full AI ecosystem available to iOS developers:

  • Foundation Models (iOS 26) — Apple's framework for accessing the on-device language model. Guided generation, structured output with Swift, no cost, no connection.
  • Core ML — Apple's framework for running ML models on-device. Compatible with models converted from PyTorch, TensorFlow and ONNX.
  • Apple Intelligence APIs — Writing Tools, Image Playground, Visual Intelligence and the system APIs that allow integrating native intelligent features.
  • Swift Concurrency — async/await, Actors and TaskGroups for managing asynchronous AI operations safely and efficiently.
  • Claude Code — We use Claude Code as a development tool to accelerate implementation, generate tests and maintain code quality.
  • Cloud APIs (Claude API, OpenAI) — Integration of cloud models when the use case requires it, with error handling, retries and streaming.

How we approach an iOS AI project

We start not with the technology, but with the problem we want to solve for the user:

  • 01. Use-case definition

    We start not with the technology, but with the problem. What intelligent feature would bring real value to the user? Does it need on-device or cloud processing? What is the expected volume? What privacy constraints exist?

  • 02. Functional prototype

    We build a rapid prototype that demonstrates technical feasibility and user experience. This allows validating the concept before investing in the full implementation.

  • 03. Production implementation

    Development of the feature with robust architecture: error handling, timeouts, fallbacks, caching, monitoring and automated testing. AI in production needs the same engineering as any critical feature.

  • 04. Optimisation and measurement

    Monitoring of usage metrics, latency, API costs and user satisfaction. We iterate on prompts, model parameters and UX based on real data.

The mobile AI opportunity window

  • The global AI app market is growing at 44.9% CAGR through 2029.
  • Gartner predicts 40% of enterprise apps will feature AI agents by end of 2026, up from under 5% in 2025.
  • Apps incorporating generative AI report 40% more engagement than conventional apps.
  • While demand for AI features explodes, the pool of native iOS developers globally has contracted. Profiles combining deep native iOS experience with the ability to integrate on-device AI are especially hard to find.

Companies we have worked with

We bring enterprise iOS development experience to on-device AI integration. The same rigour applied at Santander and AXA, now for AI-powered iOS features.

Zara
Inditex
AXA
Banco Santander
Softtek
ZEAL Network SE
Indra
ONCE
El País
Direct Seguros
B-FY
knowmad mood
WISE SECURITY
Plexus Tech

Testimonials

“ It is a pleasure to recommend Fran, who is a key member of the team at B-FY, serving as the lead for iOS development in Biocryptology. He is responsible for app programming and security libraries, showcasing profound technical knowledge and an admirable ability to deliver results of the highest quality.

He stands out for his diligence and responsibility, consistently meeting established deadlines. His work, both in implementing new features and enhancing security, is always aligned with the product goals and meticulously planned, enabling him to meet sprint objectives without fail.

When the iOS team expanded within the company, Fran took on the challenge of coordinating the team, leading both functional development and code migrations to Swift. His leadership has been crucial in maintaining the quality and cohesion of the product during a period of significant technological change. He consistently demonstrates a proactive and problem-solving attitude, offering innovative solutions and ensuring the team achieves collective success.

His work is impeccable, and his commitment to excellence is evident in every project. Beyond his technical skills, his approach fosters a collaborative and respectful team environment. I am confident that his analytical ability, meticulousness, and team leadership skills are a valuable asset to any company he chooses to join. ”

“ I am very pleased to recommend Francisco for any position in software development, especially in the iOS field. I have been fortunate to work closely with him during his time on our development team and the various new connections we have had in the workplace, where he has proven to be an exceptional professional.

He distinguishes himself by his thoroughness and commitment to quality in every task he undertakes. His work is based on Clean Architecture and SOLID principles, which is reflected in the clarity, efficiency and maintainability of his code. These practices not only ensure the quality of the final product, but also facilitate collaboration within the team and the scalability of projects.

In addition to his technical skills, Francisco has a remarkable ability to communicate his ideas clearly and concisely. This has been particularly valuable in presenting solutions and defending his proposals, always with well-founded arguments based on industry best practices.

During his time in our team, he has proven to be an invaluable asset, not only for his talent and knowledge, but also for his positive attitude and willingness to help his colleagues. His ability to lead and collaborate has contributed significantly to the success of our projects.

In summary, I consider Francisco to be an outstanding professional who will bring value to any organisation lucky enough to have him. I am sure he will continue to be a positive influence and a driver of innovation in his next job challenge. ”

“ Francisco José is highly skilled in iOS development, and it’s evident that he is truly passionate about it. He has all the necessary hardware resources and is well-versed in agile methodologies. In our case, we worked with Trello and Excel. Communication with him is seamless, which makes it easy to resolve any questions that arise during development. I would definitely recommend him. ”

“ Fran is a highly experienced professional who is not only contributing to our iOS app development but also to many other key areas essential for the growth of a start-up. ”

“ I have been working with Fran as an Acilia resource for a couple of years, and his best qualities are his proactivity, his interest in continuous learning and staying up-to-date with web and app technologies, and his ability to work in a team, collaborating with various profiles to carry out developments. ”

“ Francisco José is a great professional and very hardworking with high motivation and a positive spirit. I appreciate that he is very calm and creates a good atmosphere within the team. ”

“ Fran is a well-trained and experienced worker, always learning new things. Very methodical in everything he does and with very clear ideas. Always willing to share his knowledge with the group. And as a person, he is an excellent guy! ”

“ Francisco is a great professional, and this is evident in his more than 11 years of experience in the internet sector.

This extensive experience, along with his continuous desire for learning and training, makes Francisco a highly qualified professional for software development and the internet sector. The work with multiple technologies that he has done over all these years of experience broadens the range of challenges that Francisco can successfully tackle.

Additionally, on a human level, Francisco has an open character, is capable of delegating and sharing, skills that make him very capable of working in a team. Another remarkable aspect of Francisco is his initiative in sharing knowledge and learning from others, which fosters the learning of the entire group.

During the almost three years I worked with Francisco, he was a true reference for me, as well as the person who introduced me to the sector and trained me during that time. ”

“ An example of order, seriousness, and immense love for his work. Factors that not only affect his productivity and performance but also translate into a close relationship with his colleagues, being attentive and willing to listen, debate, teach, share… undoubtedly, his desire to progress will take him wherever he wants. ”

Loading...

Frequently asked questions about AI in iOS

Yes. It requires iPhone 15 Pro or later (A17 Pro chip), or any iPhone 16. On iPad and Mac, any device with M1 chip or later. The model requires approximately 7 GB of storage on the device.

Yes. The model runs entirely on the device. It does not need an internet connection or send data to external servers. It is the ideal option for features that handle sensitive data.

On-device for tasks requiring low latency, privacy or offline functionality: classification, short summaries, entity extraction. Cloud for tasks needing more capacity: long text generation, complex reasoning, multimodal analysis.

We design architectures with fallback: on-device model if available, cloud API as an alternative, and always a graceful degraded experience for devices without AI support.

We use Claude Code as a development tool — just like we use Xcode, Git or Instruments. It lets us iterate faster, generate tests and maintain quality. All code goes through human review before being integrated.