Swift Package IndexPackages for keyword “on-device-ai”
These packages are available as a package collection, usable in Xcode or SwiftPM.
LiteRTLM
LiteRT-LM is Google's production-ready, high-performance, open-source inference framework for deploying Large Language Models on edge devices.
Wax
Shared Single-file memory layer for all your agents, sub mili-second RAG over text, photo and video on Apple Silicon.. No Server. No API. One File. Pure Swift
AgentRunKit
Swift 6 agent SDK: type-safe tools, streaming, cloud + on-device inference via MLX on Apple Silicon
coreai-kit
Swift SDK for running chat, vision and speech models on iPhone and Mac with Apple's Core AI. Model download and caching, FoundationModels integration, and runnable examples with documented OS, SDK and model requirements.
ManifoldKit
Reusable SwiftUI chat framework for local and cloud LLM inference on Apple platforms
DProvenanceKit
Local-first AI reasoning observability for Swift: record, diff, attest, and CI-gate agent execution paths. Complements OpenTelemetry and tools like LangSmith — does not replace them.
AIChatKitMLX
Adds on-device Apple MLX inference to any app already using AIChatKit. Models are downloaded from Hugging Face Hub on first use and cached locally. Runs on Metal GPU and Apple Neural Engine — no network calls during inference.
TalaDB
TalaDB for Swift - Swift bindings for TalaDB, an embedded document and vector database, for native iOS and macOS apps .
NobodyWho
On-device AI inference for Swift apps. Run LLMs locally on iOS, macOS, visionOS and watchOS.
OpenJevSwift
Native Swift implementation of OpenJev, the Jev-compatible typed-decision server: yes/no, choice and score answers read from model probabilities, on Apple silicon Macs and in iPhone apps. Serves Verdict, Laya, DiffusionGemma and JevK5. Apache-2.0.
AIChatKitLlama
Adds on-device GGUF inference via llama.cpp to any app already using AIChatKit. Models run entirely in-process using Metal GPU acceleration — no network calls after the initial download.
PrivateFoundationModels
Apple FoundationModels API on iOS 18+. Same call site, native passthrough on iOS 26 (Apple Intelligence), CoreML / MLX backends on older OSes. Drop-in source compatible.
ContextCore
Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 60-95% fewer tokens, same answers
perso-interactive-ondevice-sdk-swift
Interactive SDK for Conversational AI Avatar on Apple platforms
VeloxQuant
Swift SDK for VeloxQuant (Apple Silicon MLX KV-cache compression) — sibling to the TS, Go, Rust, and Kotlin client SDKs.
Arbiter
Framework to talk to AI models and intelligently route
FiniteExtract
On-device structured extraction engine (Swift Package)
swift-llama
Run LLMs natively in Swift. Actor-safe. Streaming. Tool-calling. Metal-accelerated.
18 packages.