mem0ai/mem0 · error · ValueError

Invalid docstore key type: {type(key)}, expected str

Error message

Invalid docstore key type: {type(key)}, expected str

What it means

Raised while validating each entry of the loaded docstore: a key (the memory id) is not a str. Mem0 requires string ids because they are also the payload lookup keys after load. ValueError surfaced during FAISS load.

Source

Thrown at mem0/vector_stores/faiss.py:107

    Raises:
        ValueError: If the data structure is invalid.
    """
    if not isinstance(data, tuple) or len(data) != 2:
        raise ValueError("Invalid docstore format: expected tuple of (docstore, index_to_id)")

    docstore, index_to_id = data

    if not isinstance(docstore, dict):
        raise ValueError("Invalid docstore format: docstore must be a dict")

    if not isinstance(index_to_id, dict):
        raise ValueError("Invalid docstore format: index_to_id must be a dict")

    # Validate docstore entries
    for key, value in docstore.items():
        if not isinstance(key, str):
            raise ValueError(f"Invalid docstore key type: {type(key)}, expected str")
        if not isinstance(value, dict):
            raise ValueError(f"Invalid docstore value type: {type(value)}, expected dict")

    # Validate index_to_id entries
    for key, value in index_to_id.items():
        if not isinstance(key, int):
            raise ValueError(f"Invalid index_to_id key type: {type(key)}, expected int")
        if not isinstance(value, str):
            raise ValueError(f"Invalid index_to_id value type: {type(value)}, expected str")

    return docstore, index_to_id


class OutputData(BaseModel):
    id: Optional[str]  # memory id
    score: Optional[float]  # distance
    payload: Optional[Dict]  # metadata

View on GitHub (pinned to 001c235229)

Solutions

Sanitize the file: stringify all docstore keys (and matching index_to_id values) before loading
Regenerate the store through mem0's API so ids are always str(uuid4())
Delete the store and re-add memories if the data is reproducible

Example fix

# sanitize before load
import json
data = json.load(open(p))
docstore = {str(k): v for k, v in data[0].items()}
data[0] = docstore
json.dump(data, open(p, 'w'))

Defensive patterns

Strategy: validation

Validate before calling

bad = [k for k in data[0] if not isinstance(k, str)]
if bad:
    data[0] = {str(k): v for k, v in data[0].items()}
    data[1] = {i: str(vid) for i, vid in data[1].items()}

Type guard

def docstore_keys_are_str(d) -> bool:
    return isinstance(d, dict) and all(isinstance(k, str) for k in d)

Prevention

Let mem0 generate ids (str(uuid4())) instead of supplying custom non-string ids
Sanitize external data before importing into a docstore

When it happens

Trigger: A persisted docstore dict containing non-string keys, e.g. integer ids from an externally built file or JSON keys that were coerced to numbers by another tool and re-serialized via pickle.

Common situations: Importing a docstore generated outside mem0; using uuid objects as keys in a custom pickle; schema drift after a manual data merge.

Related errors

AI-assisted analysis of mem0ai/mem0@001c235229 (2026-08-15). Data as JSON: /api/errors/4893efe74cf7adb9. Report an issue: GitHub.