Why are we still sending the entire conversation to the inference server on every turn? Am I missing something obvious?

🌐 链接: https://www.reddit.com/r/LocalLLaMA/comments/1x0v61i/why_are_we_still_sending_the_entire_conversation/
🏷️ 分组: Reddit｜AI·自托管·VPS
🕒 时间: 2026-10-09 00:15:26
Reddit
From the LocalLLaMA community on Reddit
From the LocalLLaMA community on Reddit
Explore this post and more from the LocalLLaMA community