Ollama
The one-liner that made local AI normal
What it is
A local LLM runner that packages weights, configuration, and a REST API behind a Docker-like pull-and-run workflow. Runs the current generation of open-weight models — Kimi, GLM, DeepSeek, gpt-oss, Qwen, Gemma — on macOS, Linux, and Windows.
Why it's interesting
The de facto front door to local inference. Its model library and single-command UX turned 'run an LLM on your laptop' from a weekend project into a one-liner, and a whole ecosystem of UIs and tools builds against its API.
Use cases
- Local, offline inference for chat and coding
- Privacy-sensitive prototyping without cloud APIs
- A backend for local AI interfaces
Who it's for
Everyone from first-time tinkerers to teams avoiding cloud calls
Setup
Easy. One installer or shell command; RAM scales with model size (a 7B model wants ~8GB)
Limitations & cautions
Inference only, and quantized GGUF models trade some quality for size. Big models still need big hardware, no matter how friendly the CLI.
Editorial takeaway
If you recommend one tool to someone who has never run a model locally, it's still this one.