mem0ai/mem0 · error · ImportError

The 'ollama' library is required. Please install it using 'p

Error message

The 'ollama' library is required. Please install it using 'pip install ollama'.

What it means

Module-level ImportError raised when mem0/embeddings/ollama.py is imported and the `ollama` Python package is not installed. Because it fires at import time, the failure occurs when mem0 lazily loads the OllamaEmbedding class (provider: 'ollama'), not at first embed call.

Source

Thrown at mem0/embeddings/ollama.py:9

from typing import Literal, Optional

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

try:
    from ollama import Client
except ImportError:
    raise ImportError("The 'ollama' library is required. Please install it using 'pip install ollama'.")


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

        self.config.model = self.config.model or "nomic-embed-text"
        self.config.embedding_dims = self.config.embedding_dims or 512

        self.client = Client(host=self.config.ollama_base_url)
        self._ensure_model_exists()

    @staticmethod
    def _normalize_model_name(name: str) -> str:
        return name if ":" in name else f"{name}:latest"

    def _ensure_model_exists(self):
        """

View on GitHub (pinned to 001c235229)

Solutions

pip install ollama (or add it to your project dependencies)
If you use uv/poetry/pipenv, add ollama to the lockfile and reinstall
Verify with python -c "import ollama" that the package resolves in the same interpreter/venv mem0 runs in

Example fix

// before
Memory.from_config({"embedder": {"provider": "ollama"}})  # ImportError

# after
# pip install ollama
Memory.from_config({"embedder": {"provider": "ollama", "config": {"model": "nomic-embed-text"}}})

Defensive patterns

Strategy: validation

Validate before calling

import importlib.util

if importlib.util.find_spec("ollama") is None:
    raise SystemExit("Missing dependency: pip install ollama")

Memory.from_config({"embedder": {"provider": "ollama", "config": {"model": "nomic-embed-text"}}})

Try / catch

try:
    from mem0.embeddings.ollama import OllamaEmbedding
except ImportError as e:
    print(f"Dependency missing: {e}. Install with: pip install ollama")
    raise SystemExit(1) from e

Prevention

Declare ollama in requirements.txt alongside the mem0 config that uses it
Add a startup dependency check for optional providers
Use one lockfile per deployment environment

When it happens

Trigger: Configuring provider: ollama in the embedder config without pip-installing ollama; using mem0ai without the [graph] extras in an env where only core deps exist; a fresh virtualenv where ollama was never added to requirements.txt.

Common situations: New Ollama users who installed the Ollama desktop app but not the Python client; CI environments missing the dependency; dependency resolution dropping ollama after a lockfile regeneration.

Related errors

AI-assisted analysis of mem0ai/mem0@001c235229 (2026-08-15). Data as JSON: /api/errors/c054cd4acde7b2a0. Report an issue: GitHub.