vLLM – Pricing & Plans

vLLM is an open-source high-throughput inference and serving engine for deploying large language and multimodal AI models efficiently on GPUs, CPUs, and other accelerator hardware.