Tensormesh, an AI infrastructure company pioneering caching-accelerated inference optimization for enterprise AI, announced its public launch and $4.5 million seed funding round led by Laude Ventures. The company’s technology reduces redundant computation during inference, cutting GPU costs and latency by up to 10x while allowing enterprises to retain full control of their infrastructure and data.
Founded by faculty and PhD researchers from the University of Chicago, UC Berkeley, and Carnegie Mellon University, Tensormesh builds on extensive academic work in distributed systems and AI infrastructure. The company is led by Junchen Jiang, a University of Chicago faculty member and co-creator of LMCache—the leading open-source key-value caching project with more than 5,000 GitHub stars and over 100 contributors. LMCache has been integrated into frameworks such as vLLM and NVIDIA Dynamo and is used across the ecosystem by organizations including Bloomberg, Red Hat, Redis, Tencent, GMI Cloud, and WEKA.
Tensormesh is the first commercial platform to productize caching for large-scale AI inference, combining the performance innovations of LMCache with enterprise-grade usability, security, and scalability. Its platform allows organizations to optimize AI inference workloads without compromising on data control or infrastructure flexibility.
The company’s distributed caching capabilities enable sharing of key-value cache (KV-cache) across nodes, a breakthrough that drives major efficiency gains for large language model (LLM) deployments. Tensormesh supports multiple storage backends for low-latency, high-throughput inference across clusters, making it a critical layer in the enterprise AI stack.
As inference workloads surge and enterprises look for sustainable efficiency solutions, Tensormesh’s caching-based approach promises to dramatically reduce compute waste while maintaining cloud-agnostic flexibility. The company’s beta is now available to enterprises seeking to accelerate AI performance on their own infrastructure or in public clouds.
KEY QUOTES
“Enterprises today must either send their most sensitive data to third parties or hire entire engineering teams to rebuild infrastructure from scratch. Tensormesh offers a third path: run AI wherever you want, with state-of-the-art optimizations, cost savings, and performance built in.”Junchen Jiang, Founder and CEO, Tensormesh
“Enterprises everywhere are wrestling with the huge costs of AI inference. Tensormesh’s approach delivers a fundamental breakthrough in efficiency and is poised to become essential infrastructure for any company betting on AI.”
Ion Stoica, Advisor to Tensormesh and Co-Founder & Executive Chairman, Databricks
“We have closely collaborated with Tensormesh to deliver an impressive solution for distributed LLM KVCache sharing across multiple servers. Redis combined with Tensormesh delivers a scalable solution for low-latency, high-throughput LLM deployments. The benchmarks we ran together demonstrated remarkable improvements in both performance and efficiency, and we’re excited to see the Tensormesh product, which we believe will set a new bar for LLM hosting performance.”
Rowan Trollope, CEO, Redis
“Our partnership with Tensormesh and integration with LMCache played a critical role in helping WEKA open-source aspects of our breakthrough Augmented Memory Grid solution, enabling the broader AI community to tackle some of the toughest challenges in inference today.”
Callan Fox, Lead Product Manager, WEKA
“Caching is one of the most underutilized levers in AI infrastructure, and this team has found a smart, practical way to apply it at scale. This is the moment to define a critical layer in the AI stack, and Tensormesh is well positioned to own it.”
Pete Sonsini, Co-Founder and General Partner, Laude Ventures
