What causes "`model` parameter is required" in mem0ai/mem0?

Pass a Langchain Embeddings instance as the model, e.g. LangchainEmbedding(BaseEmbedderConfig(model=OpenAIEmbeddings(openai_api_key=...)))

mem0ai/mem0 · error · ValueError

`model` parameter is required

Error message

`model` parameter is required

What it means

Raised by LangchainEmbedding.__init__ when the embedder config has no `model` value. Unlike other Mem0 embedding providers that default to a hosted model name, the Langchain integration has no default: the model IS a Langchain `Embeddings` object the user must supply. The constructor fails fast at instantiation time, before any network call.

Source

Thrown at mem0/embeddings/langchain.py:17

from typing import Literal, Optional

from mem0.configs.embeddings.base import BaseEmbedderConfig
from mem0.embeddings.base import EmbeddingBase

try:
    from langchain.embeddings.base import Embeddings
except ImportError:
    raise ImportError("langchain is not installed. Please install it using `pip install langchain`")


class LangchainEmbedding(EmbeddingBase):
    def __init__(self, config: Optional[BaseEmbedderConfig] = None):
        super().__init__(config)

        if self.config.model is None:
            raise ValueError("`model` parameter is required")

        if not isinstance(self.config.model, Embeddings):
            raise ValueError("`model` must be an instance of Embeddings")

        self.langchain_model = self.config.model

    def embed(self, text, memory_action: Optional[Literal["add", "search", "update"]] = None):
        """
        Get the embedding for the given text using Langchain.

        Args:
            text (str): The text to embed.
            memory_action (optional): The type of embedding to use. Must be one of "add", "search", or "update". Defaults to None.
        Returns:
            list: The embedding vector.
        """

        return self.langchain_model.embed_query(text)

View on GitHub (pinned to 001c235229)

Solutions

Pass a Langchain Embeddings instance as the model, e.g. LangchainEmbedding(BaseEmbedderConfig(model=OpenAIEmbeddings(openai_api_key=...)))
If configuring via dict/YAML, include the instantiated Embeddings object under the model key (object, not a string name)
If you meant to use a hosted provider instead, switch the embedding provider config to openai/ollama/etc. rather than langchain

Example fix

// before
embedder = LangchainEmbedding(BaseEmbedderConfig())  # ValueError: `model` parameter is required

# after
from langchain_openai import OpenAIEmbeddings
embedder = LangchainEmbedding(BaseEmbedderConfig(model=OpenAIEmbeddings(model="text-embedding-3-small")))

Defensive patterns

Strategy: validation

Validate before calling

from mem0.configs.embeddings.base import BaseEmbedderConfig
from mem0.embeddings.langchain import LangchainEmbedding

cfg = BaseEmbedderConfig()
if getattr(cfg, "model", None) is None:
    raise SystemExit("langchain embedder requires an Embeddings instance in config.model")
embedder = LangchainEmbedding(cfg)

Type guard

from langchain.embeddings.base import Embeddings

def has_langchain_model(cfg) -> bool:
    return getattr(cfg, "model", None) is not None and isinstance(cfg.model, Embeddings)

Try / catch

try:
    embedder = LangchainEmbedding(config)
except ValueError as e:
    # config construction error: report which parameter is wrong and stop
    raise ConfigurationError(str(e)) from e

Prevention

Always construct the Embeddings object first and pass it as model
Treat langchain differently from hosted providers: model is an object, never a string
Add a config smoke test that instantiates every configured provider at startup

When it happens

Trigger: Calling LangchainEmbedding() or LangchainEmbedding(BaseEmbedderConfig()) with no `model` field, or building a config dict/YAML for Memory() that names the langchain provider but omits the model key.

Common situations: Copy-pasting a config template from another provider (e.g. openai) where model is a string default; migrating from an older mem0 version where the langchain embedder behaved differently; assuming `model` means a model name string.

Related errors

AI-assisted analysis of mem0ai/mem0@001c235229 (2026-08-15). Data as JSON: /api/errors/d0cfd95f01a1388a. Report an issue: GitHub.