How To Run DeepSeek R1 on AWS Using Infrastructure as Code

DeepSeek, a new open-source language model, offers competitive performance at lower costs than major commercial models. It achieved 79.8% accuracy on AIME 2024 mathematics tests and can be deployed on standard cloud infrastructure. Using Pulumi and AWS, a practical deployment process was created with a GPU-enabled EC2 instance, Ollama for model management, and Open WebUI for interaction. This setup allows teams to run high-performance AI models with less investment, potentially making enterprise-grade AI more accessible to smaller organizations.