Inferact
Inferact is a San Francisco-based AI infrastructure company focused on solving the growing challenges of deploying and running large AI models efficiently at scale. Founded by the creators and core maintainers of vLLM, an open-source LLM inference engine, Inferact aims to make AI inference cheaper, faster, and easier to deploy across different models and accelerator hardware. The company works at the intersection of AI models, inference software, and computing hardware. As modern models become larger and more complex—including mixture-of-experts, multimodal, and agentic architectures—Inferact focuses on improving the infrastructure required to serve them efficiently. Its work builds on the vLLM ecosystem, which the company says supports more than 500 model architectures and 200 accelerator types, with contributions from more than 2,000 developers.