LMCacheย ๐คย storage at #FMS2026
Nearly every storage vendor at FMS used LMCache to show off their KV cache offload performance this week.
This is a great example of what a common, open source KV cache layer can enable as KV cache becomes a bigger part of the AI inference stack.
That's a wrap on #FMS2026! ๐ Thanks to everyone who stopped by to connect with Tensormesh. A highlight: Yihua's panel on managing KV cache across GPUs, host memory & storage as context windows and inference workloads scale. Great conversations all around. Plus, it was great to
