Ship It

Ollama, MLX, vLLM — from notebook to production

Practical deployment. Emulated terminals. Setup simulators. By the end you'll be able to answer 'which engine and what config?' for any scenario on the map.

badge · Shipwright

0 of 3 lessons completed

1
Ollama in 60 seconds
Install Ollama, pull a model, build a Modelfile with system prompts and parameters — the fastest path from zero to local LLM inference
8 min
40 xp
2
MLX-LM on Mac
Apple's MLX framework for local LLM inference on M-series chips — unified memory, Metal acceleration, and the mlx-lm CLI
7 min
35 xp
3
vLLM in production
Tensor parallelism, continuous batching, and PagedAttention in one config — deploy a production LLM endpoint with vLLM
9 min
45 xp