Perplexity Just Open-Sourced Its Local AI Engine

Get the Agent OS & Perplexity Masterclass 👉 https://www.skool.com/ai-profit-lab-7462/about Want to make money and save time with AI? Join here: https://www.skool.com/ai-profit-lab-7462/about Get a FREE AI SEO Strategy Session: https://go.juliangoldie.com/strategy-session?utm=julian Perplexity just open-sourced Lily, the local AI engine quietly beating Apple's own tool at running models on Mac — up to 35% faster, same accuracy. Here's how it works, the real hardware requirements, and why local AI just got a lot more useful. 00:00 Hook – Software was Mac's real bottleneck, not hardware 00:34 Lily Explained – Perplexity open-sources its local AI engine 00:57 Hybrid Compute – Splits cloud + on-device tasks for privacy 02:25 Prefill vs Decode – Why Lily beats general-purpose tools 03:04 Under the Hood – 35B params, Rust, Apple Metal 03:31 Benchmarks – Up to 35% faster than Apple's own tool 04:17 The Catch – Needs 32GB RAM Apple Silicon Mac 05:07 Easy Mode – One-click model picker, no setup needed 05:58 Takeaway – Local AI is finally fast enough to matter