mem0ai/mem0 · error · ImportError
cohere package is required for CohereReranker. Install…
Error message
cohere package is required for CohereReranker. Install with: pip install cohere
What it means
CohereReranker.__init__ raises ImportError when the optional `cohere` package is absent (the module guards its import with a COHERE_AVAILABLE flag). Rerankers are optional extras in mem0, so the class is importable but not constructible without its backing SDK. The message tells you the exact install command.
Solutions
- pip install cohere in the environment that runs mem0 (add it to requirements/pyproject extras)
- Or remove/disable the reranker config if reranking is optional for you
- Verify with python -c "import cohere" against the same interpreter/venv mem0 runs under
Example fix
# before
config = {"reranker": {"provider": "cohere", "config": {"model": "rerank-v3.5", "api_key": ck}}}
memory = Memory.from_config(config) # ImportError
# after
# pip install cohere
config = {"reranker": {"provider": "cohere", "config": {"model": "rerank-v3.5", "api_key": ck}}}
memory = Memory.from_config(config) Defensive patterns
Strategy: validation
Validate before calling
import importlib.util
if importlib.util.find_spec("cohere") is None:
raise RuntimeError("install cohere before enabling the cohere reranker") Try / catch
try:
reranker = CohereReranker(cfg)
except ImportError:
reranker = None # or fall back to no reranking / another provider Prevention
- Declare optional provider deps in the same lockfile as mem0
- Add a config-driven dependency check at app startup
- Use lean prod images only after confirming which extras your config needs
When it happens
Trigger: Configuring Memory with reranker={'provider': 'cohere', ...} (or instantiating CohereReranker directly) in an environment without `pip install cohere`; a Docker/CI image built from a minimal mem0 install without extras; adding a reracker config that works on a teammate's machine but not on yours.
Common situations: Enabling reranking for better recall after following docs that assume extras installed; prod images slimmed to cut dependencies; version conflicts where cohere was removed during a dependency prune.
Related errors
- Cohere API key is required. Set COHERE_API_KEY environment…
- Cohere API key is required. Set COHERE_API_KEY environment…
- sentence-transformers package is required for…
- transformers package is required for HuggingFaceReranker…
- zeroentropy package is required for ZeroEntropyReranker…
AI-assisted analysis of mem0ai/mem0@001c235229 (2026-08-15).
Data as JSON: /api/errors/ca116b0078d72d6a.
Report an issue: GitHub.
Appendix: source
Thrown at mem0/reranker/cohere_reranker.py:27
COHERE_AVAILABLE = True
except ImportError:
COHERE_AVAILABLE = False
logger = logging.getLogger(__name__)
class CohereReranker(BaseReranker):
"""Cohere-based reranker implementation."""
def __init__(self, config):
"""
Initialize Cohere reranker.
Args:
config: CohereRerankerConfig object with configuration parameters
"""
if not COHERE_AVAILABLE:
raise ImportError("cohere package is required for CohereReranker. Install with: pip install cohere")
self.config = config
self.api_key = config.api_key or os.getenv("COHERE_API_KEY")
if not self.api_key:
raise ValueError("Cohere API key is required. Set COHERE_API_KEY environment variable or pass api_key in config.")
self.model = config.model
self.client = cohere.Client(self.api_key)
def rerank(self, query: str, documents: List[Dict[str, Any]], top_k: int = None) -> List[Dict[str, Any]]:
"""
Rerank documents using Cohere's rerank API.
Args:
query: The search query
documents: List of documents to rerank
top_k: Number of top documents to return
View on GitHub (pinned to 001c235229)