mem0 integration

[sacredvoid]

Hi! I was trying to integrate mem0 to my langfuse traced LLM app, and it's struggling to capture the internal mem0 calls made to OpenAI to get embeddings and extract facts/memories. Has anyone done this successfully? I see Langtrace has integrated mem0, I love langfuse and don't really wanna switch. :)

[jannikmaierhoefer]

Hi @sacredvoid, we do not have a native integration for this. I just saw that OpenLit also has an integration. This allows us to send OTel spans to the Langfuse OTel backend to see the basic PUT and GET requests mem0 does.

You can instrument this as follows:

import os
from langfuse import get_client

# Get keys for your project from the project settings page: https://cloud.langfuse.com
os.environ["LANGFUSE_PUBLIC_KEY"] = "pk-lf-..." 
os.environ["LANGFUSE_SECRET_KEY"] = "sk-lf-..." 
os.environ["LANGFUSE_HOST"] = "https://cloud.langfuse.com" # 🇪🇺 EU region
# os.environ["LANGFUSE_HOST"] = "https://us.cloud.langfuse.com" # 🇺🇸 US region

langfuse = get_client()

# Initialize OpenLIT instrumentation
import openlit
openlit.init(tracer=langfuse._otel_tracer, disable_batch=True)

Hope this helps!