search

zilliztech / GPTCache

Semantic cache for LLMs. Fully integrated with LangChain and llama_index.
https://gptcache.readthedocs.io
MIT License
7.06k stars 500 forks source link

[Feature]:可以缓存embeding吗 #469

Closed yuanjie-ai closed 1 year ago

yuanjie-ai commented 1 year ago

Is your feature request related to a problem? Please describe.

可以缓存embeding吗

Describe the solution you'd like.

可以缓存embeding吗

Describe an alternate solution.

可以缓存embeding吗

Anything else? (Additional Context)

可以缓存embeding吗

SimFG commented 1 year ago

No, it is not practical to cache embeddings in a similar way as complete matching caching is required. Instead, you can convert the embeddings into bytes and store them in traditional databases such as SQLite, MySQL, and others.