{"type":"post","id":"supermicro-ibm-and-nvidia-launch-shared-kv-cache-4c4928","slug":"supermicro-ibm-and-nvidia-launch-shared-kv-cache-4c4928","title":"Supermicro, IBM and NVIDIA launch shared KV-cache storage for enterprise AI inference","description":"Supermicro, IBM and NVIDIA announced a jointly validated system that stores reusable inference KV states in shared network storage rather than relying only on GPU or CPU memory.","retrieval_nugget":"Supermicro, IBM and NVIDIA announced a jointly validated system that stores reusable inference KV states in shared network storage rather than relying only on GPU or CPU memory. In the companies’ test, a 120-billion-parameter model’s time to first token was up to 56 times faster with the cache at 130,000-token prompts—a result aimed at long-context and multi-step workloads.","published_at":"2026-11-05","updated_at":"2026-11-05","record_date":"2026-11-05","date_kind":"published_at","topics":["enterprise-ai"],"entities":["Supermicro","IBM","NVIDIA","IBM Storage Scale","NVIDIA Dynamo","NVIDIA Spectrum-X","AMD EPYC 9005"],"related_patterns":["agent-economics-moves-to-completed-work"],"novelty_label":"new_fact","what_changed":"Supermicro, IBM and NVIDIA announce a new shared KV-cache storage solution for enterprise AI inference.","prior_refs":[],"source_url":"https://ir.supermicro.com/news/news-details/2026/Supermicro-Launches-KV-Cache-Solution-with-IBM-and-NVIDIA-Demonstrating-Significant-Inference-Performance-Benefits/default.aspx","source_urls":["https://ir.supermicro.com/news/news-details/2026/Supermicro-Launches-KV-Cache-Solution-with-IBM-and-NVIDIA-Demonstrating-Significant-Inference-Performance-Benefits/default.aspx"],"schema_version":"newruntime-agent-readable-v0.2","stable_id":"post:supermicro-ibm-and-nvidia-launch-shared-kv-cache-4c4928","status":"published","visuals":[],"editor":{"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]},"routes":{"html":"https://newruntime.com/posts/supermicro-ibm-and-nvidia-launch-shared-kv-cache-4c4928/","markdown":"https://newruntime.com/posts/supermicro-ibm-and-nvidia-launch-shared-kv-cache-4c4928.md","json":"https://newruntime.com/posts/supermicro-ibm-and-nvidia-launch-shared-kv-cache-4c4928.json"},"markdown_url":"https://newruntime.com/posts/supermicro-ibm-and-nvidia-launch-shared-kv-cache-4c4928.md","next_reads":[{"type":"topic","path":"/topics/enterprise-ai/","url":"https://newruntime.com/topics/enterprise-ai/","title":"Enterprise Ai","reason":"Explore the enterprise-ai topic hub.","media_type":"text/html","markdown_url":"https://newruntime.com/topics/enterprise-ai.md","json_url":"https://newruntime.com/topics/enterprise-ai.json"},{"type":"collection","path":"/posts/","url":"https://newruntime.com/posts/","title":"Field Notes","reason":"Open the latest editorial analysis.","media_type":"text/html","json_url":"https://newruntime.com/posts.json"},{"type":"collection","path":"/sources/","url":"https://newruntime.com/sources/","title":"Sources","reason":"See the original sources this archive cites.","media_type":"text/html"}]}