Unverified50% confidenceFactExact time
文本生成阶段是memory-bound问题，速度受限于内存带宽；预填充阶段是compute-bound场景，可高度并行化
1
Sources
50%
Confidence
Long-term
Relevance
8/21/2026
First Seen
Sources
Related Claims
UnverifiedThe Prefill phase is a compute-intensive operation, while the Decode phase is a memory-bandwidth-intensive operation.69% similarUnverified实际开发中通常在存储长期记忆前让大模型对对话做总结摘要，提炼核心信息后再持久化，以节省空间并提高检索精确度69% similarVerifiedLarge language model inference is divided into two phases, prefill and decode: prefill is compute-intensive and sensitive to GPU processing power, while decode is a memory-bandwidth-bound task.68% similarUnverifiedLLMs' lack of cross-session memory makes it difficult to maintain consistency in code generated across multiple iterative rounds, which can be partially compensated for through Memory Bank structured Markdown files67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/781392API
curl https://kongchang.com/api/v1/knowledge/claims/781392MCP
get_claim(id=781392)