No API Key. No Cloud. Just Mac AI (Apfel)

Apfel lets you use the free AI model already built into your Mac, completely offline, with no API key, no cloud bill, and no model download. In this video, I’ll show you how Apfel unlocks Apple’s on-device Foundation Model through a simple command-line tool, interactive chat, and an OpenAI-compatible local server running on localhost, so developers can use Mac AI directly from the terminal or inside their own code. 🔗 Relevant Links Apfel Repo - https://github.com/Arthur-Ficial/apfel Apfel - https://apfel.franzai.com/ ❤️ More about us Radically better observability stack: https://betterstack.com/ Written tutorials: https://betterstack.com/community/ Example projects: https://github.com/BetterStackHQ 📱 Socials Twitter: https://twitter.com/betterstackhq Instagram: https://www.instagram.com/betterstackhq/ TikTok: https://www.tiktok.com/@betterstack LinkedIn: https://www.linkedin.com/company/betterstack 📌 Chapters: 0:00 Free AI already built into your Mac 0:34 Apple Intelligence, Foundation Models, and the hidden local LLM 1:25 Installing Apfel with Homebrew 1:48 Running Apfel fully offline from the terminal 2:19 Using Apfel in shell workflows and saving output to files 2:39 Reviewing code with Apfel and file input 2:55 Running Apfel as an OpenAI-compatible local server 3:22 How Apfel works under the hood 4:10 Apfel vs Ollama and MLX: what’s different? 4:38 Best use cases for Apfel and local Mac AI 5:00 Where Apfel breaks: model size, reasoning, and math 5:37 Apple’s safety guardrails and model restrictions 6:15 Open source Apfel vs closed models

From the most-viewed AI uploads of the last 12 hours · wall refreshed twice daily. Playback streams from YouTube; description as published by the channel.