microsoft/semantic-kernel · error · VectorStoreModelException

Index kind '{field.index_kind}' is not supported by Azure Co

Error message

Index kind '{field.index_kind}' is not supported by Azure Cosmos DB for MongoDB.

What it means

When CosmosMongoCollection builds index definitions, each vector field's index_kind must be one Cosmos DB for MongoDB supports (IVF_FLAT, HNSW, DISK_ANN, or DEFAULT). Any other IndexKind value raises VectorStoreModelException at index-definition build time, which typically runs when the collection is created or its indexes are materialized.

Source

Thrown at python/semantic_kernel/connectors/azure_cosmos_db.py:394

                for more information.
                Other kwargs are passed to the create_collection method.
        """
        await self._get_database().create_collection(self.collection_name, **kwargs)
        await self._get_database().command(command=self._get_index_definitions(**kwargs))

    def _get_index_definitions(self, **kwargs: Any) -> dict[str, Any]:
        """Creates index definitions for the collection."""
        indexes = [
            {
                "name": f"{field.storage_name or field.name}_",
                FieldTypes.KEY: {field.storage_name or field.name: 1},
            }
            for field in self.definition.data_fields
            if field.is_indexed or field.is_full_text_indexed
        ]
        for field in self.definition.vector_fields:
            if field.index_kind not in INDEX_KIND_MAP_MONGODB:
                raise VectorStoreModelException(
                    f"Index kind '{field.index_kind}' is not supported by Azure Cosmos DB for MongoDB."
                )
            if field.distance_function not in DISTANCE_FUNCTION_MAP_MONGODB:
                raise VectorStoreModelException(
                    f"Distance function '{field.distance_function}' is not supported by Azure Cosmos DB for MongoDB."
                )
            index_name = f"{field.storage_name or field.name}_"
            index_kind = DISTANCE_FUNCTION_MAP_MONGODB[field.distance_function]
            index: dict[str, Any] = {
                "name": index_name,
                FieldTypes.KEY: {field.storage_name or field.name: "cosmosSearch"},
                "cosmosSearchOptions": {
                    "kind": index_kind,
                    "similarity": DISTANCE_FUNCTION_MAP_MONGODB[field.distance_function],
                    "dimensions": field.dimensions,
                },
            }
            match index_kind:

View on GitHub (pinned to c028a0c7dc)

Solutions

1. Set the vector field's index_kind to IndexKind.IVF_FLAT, IndexKind.HNSW, IndexKind.DISK_ANN, or IndexKind.DEFAULT.
2. Keep separate VectorStoreCollectionDefinitions for NoSQL vs MongoDB collections if metrics differ.
3. Verify the IndexKind enum value spelling matches a supported constant.

Example fix

// before
VectorStoreRecordVectorField(name="embedding", dimensions=1536, index_kind=IndexKind.FLAT)
// after
VectorStoreRecordVectorField(name="embedding", dimensions=1536, index_kind=IndexKind.HNSW)

from semantic_kernel.connectors.azure_cosmos_db import INDEX_KIND_MAP_MONGODB

bad = [f for f in definition.vector_fields if f.index_kind not in INDEX_KIND_MAP_MONGODB]
if bad:
    raise ValueError(f"Unsupported index kinds: {[(f.name, f.index_kind) for f in bad]}")

def is_supported_mongodb_index_kind(k: IndexKind) -> bool:
    return k in INDEX_KIND_MAP_MONGODB

When it happens

Trigger: Raised in CosmosMongoCollection._get_index_definitions when a vector field's field.index_kind not in INDEX_KIND_MAP_MONGODB. Triggered when the data model declares a vector field with IndexKind.FLAT, IndexKind.QUANTIZED_FLAT, or another NoSQL-only / unsupported value.

Common situations: Reusing a NoSQL-oriented data model (which uses FLAT/QUANTIZED_FLAT/DISK_ANN) against a MongoDB collection. Copying an index_kind from a tutorial for a different store. Mismatching the index kind between two collections sharing one definition.

Related errors

• Distance function '{field.distance_function}' is not support
• Vector field '{options.vector_property_name}' not found in t
• Distance function '{field.distance_function}' is not support
• Vector property type '{field.type_}' is not supported by Azu
• Failed to search the collection.

AI-assisted analysis of microsoft/semantic-kernel@c028a0c7dc (2026-08-13). Data as JSON: /api/errors/4c6b8b5585f06e71. Report an issue: GitHub.