foundation-models-on-device

Run on-device FoundationModels LLM workflows with text generation, tool calling, and snapshot streaming.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/devopstales/ai-workflow --skill foundation-models-on-device-devopstales
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: foundation-models-on-device
Source: https://github.com/devopstales/ai-workflow/tree/main/skills/foundation-models-on-device
Command: npx skills add https://github.com/devopstales/ai-workflow --skill foundation-models-on-device-devopstales

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Apple FoundationModels enables on-device LLM capabilities, allowing private, offline text generation and structured interactions without relying on cloud services.

Core Features & Use Cases

  • On-device text generation and summarization with privacy guarantees.
  • Guided generation using @Generable for structured outputs.
  • Tool calling and snapshot streaming to drive real-time UI updates.

Quick Start

Activate FoundationModels on-device in your app and start a session with a simple prompt to generate text.

Frequently Asked Questions about foundation-models-on-device

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run on-device LLM workflows on iOS without a cloud dependency?

To run on-device LLM workflows without a cloud dependency, you can use Apple's FoundationModels on iOS 26+. This enables private, offline text generation and structured interactions directly on the device.

Does FoundationModels support tool calling and structured guided generation?

Yes, FoundationModels supports tool calling to execute actions and guided generation using the @Generable directive for structured outputs. It also provides snapshot streaming to drive real-time UI updates.

What is the token context window limitation for on-device text generation?

The token context window limitation for on-device text generation is strictly 4,096 tokens. This constraint applies to all sessions using FoundationModels for private, offline workflows.

When do I need offline text generation and summarization with privacy guarantees?

You need offline text generation and summarization with privacy guarantees when processing sensitive user data without cloud connectivity. FoundationModels enables these on-device LLM capabilities locally on iOS.

What iOS version is required to activate FoundationModels for private LLM sessions?

Activating FoundationModels for private LLM sessions requires iOS 26+ at minimum. This operating system version provides the underlying on-device infrastructure needed to run offline workflows.