Run Ollama Locally on Mac: Complete Setup Guide for M1-M5 (2026)
🩺 摘要
You have a Mac and want to run AI models locally. From M1 Air to M5 Ultra, this guide covers Ollama install, model selection by RAM, VS Code integration, and performance benchmarks.
📝 详情
Why Run Ollama on Mac?
Apple Silicon Macs are surprisingly capable AI machines. With unified memory, even a base M1 MacBook Air can run 7B models.
Step 1: Install
brew install ollama
Step 2: Pick Models
7B: 8GB+ | 13B: 16GB+ | 34B: 32GB+ | 70B: 64GB+
Step 3: Run
ollama pull qwen2.5:7b ollama run qwen2.5:7b
FAQ
Q: M1 Air 8GB? A: Yes. 1.5B-3.8B models at 15-25 tok/s. Q: Neural Engine? A: No. Uses Metal (GPU). Q: Best for Mac? A: 8-16GB -> Qwen2.5-Coder:7B.
💬 评论 (0)