Tensormesh, an AI infrastructure startup, has launched from stealth with $4.5 million in seed funding led by Laude Ventures and supported by angel investor Michael Franklin, the renowned database pioneer. The company is developing a commercial version of LMCache, an open-source utility created by co-founder Yihua Cheng that has become a staple in AI inference optimization for organizations including Google and Nvidia.
Founded by Junchen Jiang and Yihua Cheng, Tensormesh aims to dramatically reduce inference costs — by up to 10x — through advanced key-value cache (KV cache) retention and reuse. Its technology enables AI models to preserve learned data between queries, improving efficiency for chatbots and agentic systems without additional GPU demand.
The funding will accelerate the product’s commercialization as Tensormesh positions itself at the forefront of AI performance optimization, offering scalable, cost-efficient infrastructure for next-generation model deployment.
