Ollama – v0.14.2-rc0
Hey AI tinkerers! 🚀 Ollama v0.14.2-rc0 just landed — and Mac users with Apple Silicon are in for a treat! 🍏
✅ MLX build instructions added to the README — now you can compile Ollama natively on M1/M2/M3 chips, bypassing Docker and getting faster, leaner local LLM inference.
MLX = Apple’s new ML framework (think PyTorch, but built for M-series). No GPU? Still rockin’ Llama 3, DeepSeek-R1, or Mistral — just smoother and snappier.
⚠️ Still a release candidate, so keep an eye out for final tweaks — but if you’re tinkering on Mac? This is your golden ticket. 🎯
Linux & Windows folks — your Ollama magic stays untouched, no worries! 💻🛠️