qdrant inline provider

Signed-off-by: Daniele Martinoli <dmartino@redhat.com>

distributions/dependencies.json

@@ -444,6 +444,7 @@
     "psycopg2-binary",
     "pymongo",
     "pypdf",
+    "qdrant-client",
     "redis",
     "requests",
     "scikit-learn",

docs/source/providers/vector_io/qdrant.md

@@ -3,7 +3,7 @@ orphan: true
 ---
 # Qdrant
 
-[Qdrant](https://qdrant.tech/documentation/) is a remote vector database provider for Llama Stack. It
+[Qdrant](https://qdrant.tech/documentation/) is a inline and remote vector database provider for Llama Stack. It 
 allows you to store and query vectors directly in memory.
 That means you'll get fast and efficient vector retrieval.
 
@@ -17,7 +17,7 @@ That means you'll get fast and efficient vector retrieval.
 To use Qdrant in your Llama Stack project, follow these steps:
 
 1. Install the necessary dependencies.
-2. Configure your Llama Stack project to use Faiss.
+2. Configure your Llama Stack project to use Qdrant.
 3. Start storing and querying vectors.
 
 ## Installation

llama_stack/providers/inline/vector_io/qdrant/__init__.py

@@ -0,0 +1,19 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from typing import Dict
+
+from llama_stack.providers.datatypes import Api, ProviderSpec
+
+from .config import QdrantVectorIOConfig
+
+
+async def get_provider_impl(config: QdrantVectorIOConfig, deps: Dict[Api, ProviderSpec]):
+    from llama_stack.providers.remote.vector_io.qdrant.qdrant import QdrantVectorIOAdapter
+
+    impl = QdrantVectorIOAdapter(config, deps[Api.inference])
+    await impl.initialize()
+    return impl

llama_stack/providers/inline/vector_io/qdrant/config.py

@@ -0,0 +1,21 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+
+from pydantic import BaseModel
+
+from llama_stack.schema_utils import json_schema_type
+
+
+@json_schema_type
+class QdrantVectorIOConfig(BaseModel):
+    path: str
+
+    @classmethod
+    def sample_run_config(cls, __distro_dir__: str) -> dict[str, any]:
+        return {
+            "path": "${env.QDRANT_PATH:~/.llama/" + __distro_dir__ + "}/" + "qdrant.db",
+        }

llama_stack/providers/registry/vector_io.py

@@ -92,6 +92,24 @@ def available_providers() -> List[ProviderSpec]:
             ),
             api_dependencies=[Api.inference],
         ),
+        remote_provider_spec(
+            api=Api.vector_io,
+            adapter=AdapterSpec(
+                adapter_type="sample",
+                pip_packages=[],
+                module="llama_stack.providers.remote.vector_io.sample",
+                config_class="llama_stack.providers.remote.vector_io.sample.SampleVectorIOConfig",
+            ),
+            api_dependencies=[],
+        ),
+        InlineProviderSpec(
+            api=Api.vector_io,
+            provider_type="inline::qdrant",
+            pip_packages=["qdrant-client"],
+            module="llama_stack.providers.inline.vector_io.qdrant",
+            config_class="llama_stack.providers.inline.vector_io.qdrant.QdrantVectorIOConfig",
+            api_dependencies=[Api.inference],
+        ),
         remote_provider_spec(
             Api.vector_io,
             AdapterSpec(

llama_stack/providers/remote/vector_io/qdrant/config.py

@@ -23,7 +23,6 @@ class QdrantVectorIOConfig(BaseModel):
     prefix: Optional[str] = None
     timeout: Optional[int] = None
     host: Optional[str] = None
-    path: Optional[str] = None
 
     @classmethod
     def sample_run_config(cls, **kwargs: Any) -> Dict[str, Any]:

llama_stack/providers/remote/vector_io/qdrant/qdrant.py

@@ -6,7 +6,7 @@
 
 import logging
 import uuid
-from typing import Any, Dict, List, Optional
+from typing import Any, Dict, List, Optional, Union
 
 from numpy.typing import NDArray
 from qdrant_client import AsyncQdrantClient, models
@@ -16,12 +16,13 @@ from llama_stack.apis.inference import InterleavedContent
 from llama_stack.apis.vector_dbs import VectorDB
 from llama_stack.apis.vector_io import Chunk, QueryChunksResponse, VectorIO
 from llama_stack.providers.datatypes import Api, VectorDBsProtocolPrivate
+from llama_stack.providers.inline.vector_io.qdrant import QdrantVectorIOConfig as InlineQdrantVectorIOConfig
 from llama_stack.providers.utils.memory.vector_store import (
     EmbeddingIndex,
     VectorDBWithIndex,
 )
 
-from .config import QdrantVectorIOConfig
+from .config import QdrantVectorIOConfig as RemoteQdrantVectorIOConfig
 
 log = logging.getLogger(__name__)
 CHUNK_ID_KEY = "_chunk_id"
@@ -99,7 +100,9 @@ class QdrantIndex(EmbeddingIndex):
 
 
 class QdrantVectorIOAdapter(VectorIO, VectorDBsProtocolPrivate):
-    def __init__(self, config: QdrantVectorIOConfig, inference_api: Api.inference) -> None:
+    def __init__(
+        self, config: Union[RemoteQdrantVectorIOConfig, InlineQdrantVectorIOConfig], inference_api: Api.inference
+    ) -> None:
         self.config = config
         self.client = AsyncQdrantClient(**self.config.model_dump(exclude_none=True))
         self.cache = {}