🚀 Perplexity just open‑sourced Lily, a Rust‑powered inference engine with Metal kernels for Qwen 3.6‑35B on Apple Silicon. Run local, mix‑of‑experts models fast. Dive in to see how MLX‑LM gets a boost! #LilyEngine #Qwen3_6_35B #AppleSilicon
@aidailypost.com