mem0ai/mem0 · error · ValueError

Cohere API key is required. Set COHERE_API_KEY environment v

Error message

Cohere API key is required. Set COHERE_API_KEY environment variable or pass api_key in config.

What it means

CohereReranker requires an API key: config.api_key or the COHERE_API_KEY environment variable; if both are empty it raises ValueError at construction. The key is needed to build cohere.Client for the rerank API. This fails before any network call, at Memory.from_config()/init time.

Source

Thrown at mem0/reranker/cohere_reranker.py:32


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
            
        Returns:
            List of reranked documents with rerank_score
        """
        if not documents:
            return documents

View on GitHub (pinned to 001c235229)

Solutions

Export COHERE_API_KEY in the environment where mem0 runs, or pass config: {'provider':'cohere','config':{'api_key': ...}}
For containers/CI, inject the variable at deploy time and verify with print(os.getenv('COHERE_API_KEY')) inside the process
Never commit the key; use env/secret managers

Example fix

# before
config = {"reranker": {"provider": "cohere", "config": {"model": "rerank-v3.5"}}}
# no COHERE_API_KEY set -> ValueError

# after
import os
os.environ["COHERE_API_KEY"] = os.environ["COHERE_API_KEY"]  # from secret manager
config = {"reranker": {"provider": "cohere", "config": {"model": "rerank-v3.5", "api_key": os.environ["COHERE_API_KEY"]}}}

Defensive patterns

Strategy: validation

Validate before calling

import os
cfg = {"provider": "cohere", "config": {"api_key": os.getenv("COHERE_API_KEY")}}
if not cfg["config"]["api_key"]:
    raise RuntimeError("COHERE_API_KEY not set; refusing to build cohere reranker")

Prevention

Fail fast at startup on missing secrets rather than at Memory construction
Inject secrets via env/secret manager in every environment including CI
Never default api_key to empty string; make absence loud

When it happens

Trigger: Configuring the cohere reranker with no api_key in config and no COHERE_API_KEY exported; setting the var in your shell but not in the Docker container/systemd unit/CI where the process actually runs; config.api_key set to an empty string (falsy) with the env var also unset.

Common situations: Local config works, deployed environment lacks the env var; .env file not loaded by the service; secrets moved to a vault but the reranker config never updated; empty-string defaults from template configs.

Related errors

AI-assisted analysis of mem0ai/mem0@001c235229 (2026-08-15). Data as JSON: /api/errors/bf9fa6e8bed878a3. Report an issue: GitHub.