🚀 𝐋𝐌𝐂𝐚𝐜𝐡𝐞 𝐯𝟎.𝟒.𝟕 𝐢𝐬 𝐡𝐞𝐫𝐞! Release highlights: 🧠 Wider model support — LMCache now supports hybrid models including DeepSeek V4, Gemma, and Qwen 3.5/3.6 (Mamba/GDN), via per-group block sizes that handle their KV formats. 💻 Run LMCache on CPU-only machines! — A new shared-memory (SHM) transfer path for MP mode lets LMCache run across GPUs, CPUs, and other accelerators, including CPU-only deployments. 🌐 The new MP coordinator provides a global control plane: server registration, CLI, L2 quota/eviction, and a fleet-wide CacheBlend fingerprint directory. ➕ New backends: NIXL DOCA_MEMOS, Google Cloud Bigtable, Moore Threads MUSA. Huge thanks to all contributors who made this release possible! 𝐑𝐞𝐚𝐝 𝐦𝐨𝐫𝐞 𝐢𝐧 𝐭𝐡𝐞 𝐫𝐞𝐥𝐞𝐚𝐬𝐞 𝐧𝐨𝐭𝐞𝐬: github.com/LMCache/LMCache/r… #LLM #AIInfrastructure #KVCache #LMCache

🎙️Tensormesh: From Research to $20M Round Our CEO & Co-Founder @JunchenJiang sat down with TechBeats pod to talk KV cache, the "Big Data of AI," and how Tensormesh became the first caching-accelerated inference platform for enterprises across the GPU ecosystem. Watch Full interview👇 youtu.be/kNoVF1p5xTA #LLMInference #KVCache #AIInfrastructure

Running open-source models is easy. Running them well in production isn’t. In today’s spotlight, Nick Barcet (@nijaba), Head of GTM @tensormesh, shares what Tensormesh is built to solve — and the 𝗽𝗲𝗼𝗽𝗹𝗲 and culture behind the execution. If you’re running OSS models and want dramatically more efficient inference 🎯Try @tensormesh with $100 in free GPU credits: lnkd.in/g-gXYtaV #AI #LLM #Inference #kvcache

WOOT! #LMCache in the CNCF Technology Radar. cncf.io/reports/cncf-technol… That's golden to our community and everyone @tensormesh #kubecon #cncf #AI #LLM #inference #Tensormesh

The result of joint work with Cohere, CoreWeave and Tensormesh is impressive. Storing kv cache for #RAG on CoreWeave AI Object Storage does bring some great improvements to the Cohere North platform. blog.lmcache.ai/2025-10-29-c… #lmcache #tensormesh #kvcache #llm #vllm #inference

💰Want CHEAP GPU cloud? 💡Wanna store ALL your users’ history&docs as KV cache to save cost—but can’t get open source to run? Try TensorMesh SaaS: ⚡️$3.09/hr H100 😄No vendor lock-in 🧠Any open-source model 🪄OpenAI-API compatible Join the waitlist🚀: tensormesh.ai/beta-waitlist

Do you want to compare the caching performance of your LLM serving stack? We've put together a simple command line tool to do so. Introducing Tensormesh Benchmark. tensormesh.ai/blog-posts/ten… #llm #ai #kvcache #lmcache #vllm #benchmarking

Tensormesh is recruiting an Account Exec and an AI Software Engineer (more positions to come soon). Check them out at lnkd.in/eaeKy4iM #AI #recruiting #job #hiring #startup

Interviewing 100 Bay Area Startups has always been my dream — and today, I’m starting the journey. 🚀 Big thanks to @lmcache for inviting me to their lunch meeting and letting me do my first interview. Great team with a great vision — they deserve more exposure. If you’re building something exciting in the Bay Area and want to share your story, let’s chat! @JunchenJiang @nijaba

I'm not coming here that much anymore. Looks like this is my X social media platform...

Hey, @VW, I reside in the US. Rented an ID.4 while in Europe. Here is what happens when I try to install We Connect ID. Normal?

#tallshipsstpete does not let you bring water in the expo and only sells single use plastic water bottles. Way to go! The ocean thanks you

What a better to launch a new product than to be able to talk about it with an actual user of the solution? Thanks to @LockheedMartin for the help in defining and implementing Red Hat Device Edge. youtu.be/fNU-3Bh_8tg @RedHat #openshift #microshift #Kubernetes #KubeCon #edge

Is it just me or do you feel like this part of the #Kubecon2022 keynote feels like watching commercials on TV?

What? It's already been 10 years since I joined eNovance and 8 since Red Hat acquired us?

6 month later, that little space experiment is now officially a product! Congratulations

Finally the word is out. So much work, but also a ton of fun creating @microshift_io with my team. I'm very grateful to work with so many talented engineers across Red Hat, from our partners, and from the community who contributed. #opensource dev model FTW! @RedHat @redhatopen

Sdxcentral: Red Hat Shrinks OpenShift for Device Edge - Red Hat Device Edge sdxcentral.com/articles/news… #rhel #openshift #redhat #edge #redhatdeviceedge #microshift

SiliconAngle: Red Hat targets complex edge uses with lightweight OpenShift implementation -Red Hat Device Edge siliconangle.com/2022/10/25/…

I'll obviously be @ #Kubecon next week. Here is what #RedHat is doing overall: red.ht/KubeConNA2022