Tensormesh has officially launched with $4.5 million in seed funding led by Laude Ventures. The company is bringing commercial-grade caching technology to enterprise AI infrastructure, promising to cut inference costs and latency by up to 10x. Their approach eliminates redundant computation in AI inference while keeping data and infrastructure under full enterprise control.

The funding round positions Tensormesh to scale its technology beyond the open-source community into mainstream enterprise adoption. The company emerged from stealth with a beta product already available, building on years of distributed systems research from top-tier universities.

Tackling the AI Infrastructure Challenge

As AI workloads grow, companies face a difficult choice: send sensitive data to third-party providers or build costly infrastructure in-house. Tensormesh offers an alternative by letting organizations run optimized AI inference on their own infrastructure, whether that’s public cloud, private data centers, or hybrid environments.

“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 of Tensormesh

The platform’s cloud-agnostic design means teams can start small and expand across any environment. It’s available both as SaaS and standalone software, giving companies flexibility in how they deploy and scale.

From Open Source to Enterprise Product

Tensormesh builds on LMCache, an open-source KV caching project with over 5,000 GitHub stars and more than 100 contributors. The project has already been integrated into major frameworks like vLLM and NVIDIA Dynamo, with users including Bloomberg, Red Hat, Redis, Tencent, GMI Cloud, and WEKA.

Junchen Jiang, a University of Chicago faculty member, co-created LMCache before founding Tensormesh. The founding team includes PhD researchers from UC Berkeley and Carnegie Mellon, bringing deep expertise in distributed systems and AI infrastructure. Tensormesh is now the first commercial platform to productize caching technology for large-scale AI inference.

The company combines LMCache-inspired techniques with enterprise features like enhanced security, usability, and management tools. This approach bridges the gap between academic research and production-ready software that enterprises can deploy immediately.

Real-World Performance Gains

Distributed KV-cache sharing across cluster nodes drives the platform’s throughput improvements and cost reductions. Tensormesh supports multiple storage backends to enable low-latency, high-throughput deployments at scale.

“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,” said Rowan Trollope, CEO of Redis.

WEKA has also integrated LMCache into its infrastructure. “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,” said Callan Fox, Lead Product Manager at WEKA.

Investor Confidence in Infrastructure Innovation

“Enterprises everywhere are wrestling with the huge costs of AI inference,” said Ion Stoica, advisor to Tensormesh and Co-Founder and Executive Chairman of Databricks. “Tensormesh’s approach delivers a fundamental breakthrough in efficiency and is poised to become essential infrastructure for any company betting on AI.”

Laude Ventures led the seed round, seeing an opportunity to establish a new infrastructure layer in the AI stack. “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,” said Pete Sonsini, Co-Founder and General Partner at Laude Ventures. “This is the moment to define a critical layer in the AI stack, and Tensormesh is well positioned to own it.”

Origianl Creator: Ekaterina Pisareva
Original Link: https://justainews.com/companies/funding-news/tensormesh-raises-4-5m-to-cut-ai-inference-costs-by-10x/
Originally Posted: Thu, 23 Oct 2025 21:24:24 +0000