Unverified50% confidenceFactExact time
Improved /fork so that new background sessions retain the prompt cache from the original conversation.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/11/2026
First Seen
Valid until: 12/10/2026
Sources
Related Entities
Related Claims
UnverifiedIn multi-turn conversations, all previous context is resent as input each time, so redundant code generated early on continues to incur token costs in every subsequent round.68% similarUnverifiedCodex supports a Fork Chat feature that allows multiple independently evolving conversation branches to be created on the basis of a shared historical context, for parallel handling of different tasks.67% similarUnverifiedPrefix caching leverages the fact that the system prompt and conversation history remain unchanged across turns in long conversations, saving the KV cache of the fixed prefix for reuse in new turns to avoid redundant computation.65% similarUnverified上下文缓存机制对多轮对话、长文档处理和批量任务等场景尤为有利65% similarUnverified用户可通过精简提示词、控制对话长度、及时开启新对话清空历史上下文来降低token消耗65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/897889API
curl https://kongchang.com/api/v1/knowledge/claims/897889MCP
get_claim(id=897889)