A garbage collector for LLM context windows, written in Rust
Hi HN, Like many of you, I spend a lot of time coding with Codex and switching between cloud and local models. As a software engineer, something I never understood, and honestly hated—was `/compact` and the way it’s designed. On local models my GPU screams, and after a few hours prefill takes 2 to 3 minutes per turn. On cloud models, as the context grows, the model starts getting noticeably dumber. IMHO the root problem is context bloat and the fact that compaction destroys the quality of the mo
阅读原始来源
来源记录
指标快照
points · 1 votes
2026-11-09 12:45:29.470Z
cumulative since publication
研究分析
来源记录
来源描述：Hi HN, Like many of you, I spend a lot of time coding with Codex and switching between cloud and local models. As a software engineer, something I never understood, and honestly hated—was `/compact` and the way it’s designed. On local models my GPU screams, and after a few hours prefill takes 2 to 3 minutes per turn. On cloud models, as the context grows, the model starts getting noticeably dumber. IMHO the root problem is context bloat and the fact that compaction destroys the quality of the mo
现有背景
来源票数: 1 votes (2026年11月9日 UTC; cumulative since publication)；前次同口径记录为 1（2026年11月9日 UTC），两个观测值不等于需求增长。 可供比较的近期同类记录：“[推广] [Sidrune AI 中转站双十一] 留 UID 送 3U｜抽奖 2000R，充值加赠最高$1000” (V2EX, 2026年11月9日)；“Automattic debuts a platform to deploy vibe-coding apps amid a changing internet” (TechCrunch, 2026年11月9日)。主题相近仅用于导航，未核验其是否属于同一需求或独立证据。
研究状态
这条记录尚不能可靠归入主类，未生成商业或性能结论。
缺失证据
需要核验主要事件、受影响对象与原始证据。
下一步
阅读原文，明确它支持什么决策，再选择对应分析框架。