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
I use Azure AI Search for mem0 and it's working great, however the mem0.memory.Memory client is syncronous and is blocking my flow adding latency. It seems the AsyncMemroyClient is only for the hosted version which I cannot use at my company. Azure AI Search can be async and I could make a PR to add it but there would need to be an AsyncMemory class in mem0.memory.main.py to use this. This situation seems like a common use case so I'm wondering if it's planned or there is some other reason it does not already exist?
Motivation, pitch
Async is critical for high concurrency and scalability. There should be an option to choose Sync/Async for self-hosted memory.
The text was updated successfully, but these errors were encountered:
🚀 The feature
I use Azure AI Search for mem0 and it's working great, however the
mem0.memory.Memory
client is syncronous and is blocking my flow adding latency. It seems the AsyncMemroyClient is only for the hosted version which I cannot use at my company. Azure AI Search can be async and I could make a PR to add it but there would need to be an AsyncMemory class in mem0.memory.main.py to use this. This situation seems like a common use case so I'm wondering if it's planned or there is some other reason it does not already exist?Motivation, pitch
Async is critical for high concurrency and scalability. There should be an option to choose Sync/Async for self-hosted memory.
The text was updated successfully, but these errors were encountered: