The vLLM Team Launches Startup Inferact, Aiming to Build a World-Class AI Inference Engine
2026-01-23 / Read about 0 minute
Author:小编   

Inferact, a startup in the AI infrastructure space, co-founded by the original creators and core maintainers of vLLM, has secured $150 million in its latest seed funding round. This impressive achievement comes with a company valuation of approximately $800 million. The funding round was co-led by prominent venture capital firms Andreessen Horowitz and Lightspeed Venture Partners, with additional participation from several well-known investment institutions.

vLLM is renowned for its support of a wide range of model architectures and accelerators, offering inference capabilities on a global scale. Building on this foundation, Inferact aims to elevate vLLM into a premier AI inference engine, thereby accelerating the progress of AI technology.

With the rising popularity of large-scale models and generative AI, there has been a significant spike in the demand for computing resources and associated costs during the inference phase. Consequently, investors are now focusing on technologies that can enhance model operational efficiency and lower inference expenses. Inferact directly addresses this critical industry need by dedicating itself to providing efficient AI inference solutions.

Recently, there has been a surge in funding for AI infrastructure companies, a trend that underscores the industry's heightened attention to inference technology. Inferact's successful funding round not only underscores the commercial potential of open-source technology but also signals a broader shift in the AI infrastructure market—from a focus on training to an emphasis on inference and deployment.