Open Models Change The Economics of AI

Open-source AI models are rapidly transforming enterprise workflows by offering cost-effective, customizable alternatives to closed frontier models. Coding agents currently drive the majority of token consumption, with organizations like AT&T shifting significant portions of their workloads to open-source solutions. Ollama facilitates this transition by serving as a critical infrastructure layer that integrates hardware, inference engines, and application runtimes, enabling developers to deploy models efficiently across local and cloud environments. While security and safety remain primary concerns for enterprise adoption, the performance gap between open and closed models is closing, with local hardware advancements allowing for high-performance execution of 20B to 120B parameter models. Future enterprise strategies will likely rely on a hybrid approach, utilizing smaller, specialized open models for routine tasks while reserving powerful frontier models for complex, high-stakes orchestration and coordination.