Skip to main content

​
Overview

Mem0 OSS can use TokenLab through its OpenAI-compatible LLM configuration by keeping provider: "openai" and setting openai_base_url.
Type: Memory FrameworkPrimary Path: OpenAI-compatible LLM configurationSupport Confidence: Supported custom endpoint path

​
Environment

export TOKENLAB_API_KEY="sk-your-tokenlab-key"

​
Python Configuration

import os

from mem0 import Memory

config = {
    "llm": {
        "provider": "openai",
        "config": {
            "model": "claude-sonnet-5",
            "api_key": os.environ["TOKENLAB_API_KEY"],
            "openai_base_url": "https://api.tokenlab.sh/v1",
            "temperature": 0.1,
        },
    },
    "vector_store": {
        "provider": "qdrant",
        "config": {"host": "localhost", "port": 6333},
    },
}

memory = Memory.from_config(config)

​
Basic Usage

memory.add("The user prefers concise engineering summaries.", user_id="demo")
results = memory.search("How should I answer?", user_id="demo")
print(results)

​
Endpoint Notes

This configuration covers Mem0’s OpenAI-compatible LLM path. Configure embedders separately according to the vector store and embedding model you use.