christopherkarani/Swarm

Type-safe tools, on-device inference, multi-agent workflows, memory, and guardrails, in one Swift-native runtime

What it solves

Swarm is a Swift-native agent runtime designed to help developers build production-ready AI agents for iOS, macOS, and Linux. It provides a type-safe way to integrate tools, manage memory, and orchestrate complex agent workflows while ensuring strict concurrency safety using Swift 6.2.

How it works

Swarm uses a pluggable InferenceProvider system, allowing agents to run on Apple Foundation Models for on-device inference or connect to external providers like Ollama, LM Studio, OpenAI, Azure, and OpenRouter. It employs a @Tool macro to generate tool schemas from Swift types at compile time, ensuring type safety. Agents can be composed into Workflow objects to create sequential, parallel, or routed execution paths.

Who it’s for

Swift developers building AI applications for Apple platforms or Linux who need a robust framework for managing agentic behavior, tool use, and on-device ML.

Highlights

  • Multi-platform support: Works across iOS, macOS, and Linux.
  • On-device inference: Native integration with Apple Foundation Models.
  • Composable workflows: Support for sequential, parallel, and routed agent chains.
  • Type-safe tools: Compile-time schema generation via Swift macros.
  • Production features: Includes guardrails, streaming, durable execution (checkpoint/resume), and OpenTelemetry observability.
  • MCP support: Ability to discover and bridge Model Context Protocol tools.

Related

  • Project
    eastriverlee/LLM.swift

    A Swift library for running local LLMs on Apple platforms, providing high-level abstractions for structured output, tool use, and chain-of-thought reasoning.

  • Project
    2FastLabs/agent-squad

    A flexible, lightweight framework for orchestrating multiple AI agents across Python, TypeScript, and Swift, featuring intelligent routing and on-device execution for Apple platforms.

  • Project
    ml-explore/mlx-swift-lm

    A Swift package for building applications with large language models and vision language models using MLX Swift, featuring model loading, fine-tuning, and guided generation.

  • Project
    kyegomez/swarms

    A production-ready multi-agent orchestration framework that provides pre-built architectures for coordinating multiple AI agents to solve complex tasks.

  • Project
    joelhooks/swarm-tools

    A multi-agent coordination framework that decomposes complex tasks into parallel subtasks with git-backed task tracking and semantic memory to prevent context loss.