You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Closing this as part of the EverOS 1.0 issue triage. This issue references the pre-1.0 API or retired infrastructure such as /api/v1/memories, /api/v3/agentic/*, MongoDB, Elasticsearch, Milvus, Redis, Kafka, longjob, or old memory type names. EverOS 1.0 uses POST /api/v1/memory/{add,flush,search,get} with Markdown + SQLite + LanceDB. Migration notes are tracked in PR #258: #258. If the same behavior still occurs on current main with the 1.0 API, please open a fresh issue with a 1.0 repro.
Activity
hl123-123 commentedon Nov 17, 2025
ES 既可以做关键词检索，又可以进行向量化检索。为啥不都用ES。
freshpomelo commentedon Nov 17, 2025
ES并不是个专业的向量数据库，是基于倒排索引机制上构建的向量搜索能力，性能、灵活性都不如milvus。这个项目也没有历史包袱，所以选用性能最好的
hl123-123 commentedon Nov 17, 2025
相对应的 舍 就是需要存储2遍对吧
freshpomelo commentedon Nov 17, 2025
向量化后的数据不论用ES还是milvus都得存，但有些公共数据是需要存储2遍。但存储效率不是当前技术选型的首要考虑
hl123-123 commentedon Nov 17, 2025
我有另一个方面的想法。就是从membase，然后构造多层次的mem系统，整个token开销，会很高。尤其构造多层的记忆系统，进行多type管理。
就是这个成本对于普通的游客用户来说是不是太大了，哪怕使用便宜的模型
hl123-123 commentedon Nov 17, 2025
或许感觉mem也可以用于 某个领域持续性的追踪，这种tob的可能 整个迁移逻辑我个人感觉挺平滑
freshpomelo commentedon Nov 18, 2025
我觉得因为要构造多层次的mem，但现在大模型性能又不太支持一次性把多层次的东西总结提取出来，就导致会多次提取。如果之后大模型性能上来了，自然开销也就下来了。
但如果额外加了记忆类型提取的可选项，可控性的确也变更高了
cyfyifanchen commentedon Jun 6, 2026
Closing this as part of the EverOS 1.0 issue triage. This issue references the pre-1.0 API or retired infrastructure such as /api/v1/memories, /api/v3/agentic/*, MongoDB, Elasticsearch, Milvus, Redis, Kafka, longjob, or old memory type names. EverOS 1.0 uses POST /api/v1/memory/{add,flush,search,get} with Markdown + SQLite + LanceDB. Migration notes are tracked in PR #258: #258. If the same behavior still occurs on current main with the 1.0 API, please open a fresh issue with a 1.0 repro.
🐛 fix(cascade): harden backfill preflight, Ctrl-C, sync scope, concur…
fix(cascade): close 5 backfill regressions and idempotency gaps
fix(component): unify embedding accessor + log config errors