MacPaw's Elix claims fastest LLM inference on Apple silicon, winning all 33 benchmarks
JosephJacks_ · x · 2026-11-11
After its latest update, MacPaw's Elix is billed as the fastest LLM inference engine on Apple silicon. Across 5 Macs and 33 chip×model runs with 7 LiquidAI models at 4-bit vs mlx-lm, mlx-vlm, mlx-serve, vLLM-Metal, llama.cpp, and uzu, Elix was end-to-end fastest in 33/33: decode +37% over the field, +43% at 32K context.
Elix is an on-device agentic AI runtime with three parts: declarative Swift AI pipelines compiled to a Codable AST with dependency analysis and parallel scheduling, a shared AI service across apps, and an MLX-based engine tuned per-chip with custom Metal kernels. Still in development, demo by request.
More from coding & agent
- Supabase lands in Expo EAS CLI: one command wires a backend into every environment — dshukertjr · 2026-11-11
- What would you rewrite with AI? Orosz sparks a list of cheaper rebuilds of existing services — _philschmid · 2026-11-11
- Ex-Meta researcher predicts a hard ceiling of 150-200B tokens for the SWE market — DimitrisPapail · 2026-11-11
- ComfyUI Challenge Winner ComfyArt: Open-Source Fully Local AI Design Canvas — bfl_ai · 2026-11-11
- Letting an agent self-improve its own harness daily: resumed runs 7→77, memory noise 94%→0 — AccBalanced · 2026-11-11
- One Photo of a Toddler's Playroom Becomes a Racing Game, Then the Whole House — every · 2026-11-11