请问，如果前期用api的形式使用，后期需要部署在本地，那之前在api线上的记忆能够迁移出来吗？ #1189
Answered
by
CaralHsi
lancecheney
asked this question in
Q&A
-
|
使用api的话，记忆我看好像是存在云端的。是否可以迁移呢 |
Beta Was this translation helpful? Give feedback.
Answered by
CaralHsi
May 12, 2026
Replies: 1 comment
-
|
可以的。您可以通过 get_memory 接口把之前在线 API 中的记忆导出出来，然后在本地环境重新导入使用。需要注意的是，迁移到本地后通常需要重新进行 embedding 操作，因为本地部署使用的向量库、embedding 模型或索引环境可能和线上 API 不完全一致。 Yes, it is possible. You can use the get_memory API to export the memories stored in the online API service, then import them into your local deployment. Please note that the memories usually need to be embedded again locally, since the local vector database, embedding model, or index environment may differ from the online API setup. |
Beta Was this translation helpful? Give feedback.
0 replies
Answer selected by
CaralHsi
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
可以的。您可以通过 get_memory 接口把之前在线 API 中的记忆导出出来，然后在本地环境重新导入使用。需要注意的是，迁移到本地后通常需要重新进行 embedding 操作，因为本地部署使用的向量库、embedding 模型或索引环境可能和线上 API 不完全一致。
Yes, it is possible. You can use the get_memory API to export the memories stored in the online API service, then import them into your local deployment. Please note that the memories usually need to be embedded again locally, since the local vector database, embedding model, or index environment may differ from the online API setup.