Unverified50% confidenceSolutionExact time
SGLang introduces a session-reference-aware unified Radix cache where requests can carry a session_id so that cache eviction recognizes prefixes still referenced by active sessions, with references released via /close_session.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/11/2026
First Seen
Valid until: 12/10/2026
Sources
Related Entities
Related Claims
UnverifiedSGLang的核心创新RadixAttention将KV缓存组织为前缀树结构，使具有公共前缀的多个请求能够共享缓存66% similarUnverifiedSGLang's RadixAttention technology uses a radix tree structure to cache shared prompt prefixes across different requests, so when 100 users use the same system prompt, the KV Cache only needs to be computed once.62% similarUnverified持久记忆系统本质上是RAG范式在对话历史领域的专项深化，检索源从静态知识库变为动态累积的用户交互记录59% similarUnverifiedRadixAttention 通过将 KV Cache 组织为前缀树结构，使具有相同前缀的多个请求共享同一份缓存57% similarUnverifiedSGLang出自伯克利，引入RadixAttention实现更高效的前缀共享和缓存管理，对Agent类工作负载优化更到位57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/897572API
curl https://kongchang.com/api/v1/knowledge/claims/897572MCP
get_claim(id=897572)