News Summary

Tensormesh, a startup focused on maximizing the efficiency of AI server loads, has raised $4.5 million in seed funding. The round was led by Laude Ventures with participation from angel investor Michael Franklin. The funds will be used to commercialize Tensormesh’s open-source LMCache utility. This technology can reduce inference costs by up to 10 times by retaining the key-value cache (KV cache) instead of discarding it after each query, a method which enhances performance particularly in chat interfaces and agentic systems. As AI companies face increasing complexity, Tensormesh offers a viable out-of-the-box solution to improve efficiency without the need for extensive in-house development.