Unverified55% confidenceFactExact time
PD分离架构允许不同阶段选用不同规格的GPU实例，并可独立扩缩Prefill或Decode节点数量
1
Sources
55%
Confidence
Long-term
Relevance
5/21/2026
First Seen
Sources
Tutorial: Deploying a PD-Disaggregated SGLang Multi-Node Inference Cluster on AMD GPUs
twitterlmsysorg5/21/2026
Related Entities
Related Claims
UnverifiedPD分离架构通过物理隔离消除Prefill对Decode的干扰，显著改善TTFT的P99延迟69% similarUnverifiedSome of the latest systems, such as Sarathi-Serve and Splitwise, have proposed a Prefill-Decode disaggregation architecture that schedules the prefill and decode stages to different GPU clusters.68% similarUnverifiedPDC mode allows models to combine multi-step operations using TypeScript programs through the CodeMod SDK, making it suitable for complex long-workflow tasks to be completed in one pass.68% similarUnverifiedPD分离式推理在pp_size>1（多阶段流水线并行）场景下获得支持67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/556API
curl https://kongchang.com/api/v1/knowledge/claims/556MCP
get_claim(id=556)