ai-agent

Llama-macOS: Run LLMs Locally on macOS with Swift Ease

Name: Llama-macOS: Run LLMs Locally on macOS with Swift Ease
Author: ggml-org

Launch your own large language models directly on macOS, perfect for founders needing secure, offline AI—built in Swift and backed by 1.5k+ GitHub stars.

From the repo · “A cosy home for your LLMs.”

1,661 stars+43★ this week130 forksSwiftGuide quality 8/10 · reviewed Oct 6Updated 10/20/2026100% free · open source

What it does

Llama-macOS provides a local environment for running and managing large language models (LLMs) on macOS devices, allowing founders to develop and test AI solutions natively.

When to use it

•When you need to develop and test AI-powered applications on macOS without relying on cloud services
•When you require a local environment for training and fine-tuning LLMs for specific business needs
•When you want to integrate LLM capabilities into your macOS-based startup products

Ready-to-paste prompt

swift run Llama-macOS --model-name 'ggml-bert-base' --input-text 'What is the meaning of AI?'

Heads up: Before running Llama-macOS, ensure you have the latest version of Xcode and the Swift compiler installed, as the project relies on specific Swift features and macOS frameworks

Saves to your device

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🛠️ Technical setup

Expect 20–40 minutes in a terminal — or let your AI agent drive it.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/llama-macos && curl -fsSL https://workflowstacks.com/api/skills/llama-macos/claude-skill -o ~/.claude/skills/llama-macos/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

How Llama-macOS: Run LLMs Locally on macOS with Swift Ease works

Codeflow

Free to inspect

Llama-macOS: Run LLMs Locally on macOS with Swift Ease is a large Swift project (~18k lines across 89 code files). It is a full software project: use it through its install path rather than reading it end to end. Last commit this month, MIT license, no tests found.

Size

Large codebase

~18k lines · 89 code files · ~3 h to skim

Setup

Developer setup

A real software project. Use it via its install path; don't expect to read it all.

Runs on

Swift

No API keys detected

Swift 95%HTML 5%

Where to start reading

1
readme.md
Start here — what it does and how to install it

What's in each folder

Llama/Server/Backend / API9 files

Llama/Assets.xcassets/Folder38 files

Llama/Menu/Folder23 files

Llama/Common/Folder8 files

Llama/Models/Data models & types8 files

Llama/HF/Folder7 files

Llama/Settings/Configuration7 files

Llama/Downloads/Folder6 files

READMENo tests foundThin docsNo CIMIT licenseUpdated this month