phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-15 18:23:09 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

chore: Updating how default embedding model is set in stack

Browse source 
Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>

# Conflicts:
#	.github/workflows/integration-vector-io-tests.yml
#	llama_stack/distributions/ci-tests/run.yaml
#	llama_stack/distributions/starter-gpu/run.yaml
#	llama_stack/distributions/starter/run.yaml
#	llama_stack/distributions/template.py
#	llama_stack/providers/utils/memory/openai_vector_store_mixin.py
This commit is contained in:
Francisco Javier Arceo 2025-10-15 17:15:43 -04:00
parent cd152f4240
commit 24a1430c8b
32 changed files with 276 additions and 265 deletions
Download patch file Download diff file Expand all files Collapse all files

10
llama_stack/providers/remote/vector_io/chroma/__init__.py
View file

@ -4,19 +4,27 @@
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
from llama_stack.core.datatypes import StackRunConfig
from llama_stack.providers.datatypes import Api, ProviderSpec
from .config import ChromaVectorIOConfig
async def get_adapter_impl(config: ChromaVectorIOConfig, deps: dict[Api, ProviderSpec]):
async def get_adapter_impl(
config: ChromaVectorIOConfig, deps: dict[Api, ProviderSpec], run_config: StackRunConfig | None = None
):
from .chroma import ChromaVectorIOAdapter
vector_stores_config = None
if run_config and run_config.vector_stores:
vector_stores_config = run_config.vector_stores
impl = ChromaVectorIOAdapter(
config,
deps[Api.inference],
deps[Api.models],
deps.get(Api.files),
vector_stores_config,
)
await impl.initialize()
return impl

12
llama_stack/providers/remote/vector_io/chroma/chroma.py
View file

@ -12,15 +12,17 @@ import chromadb
from numpy.typing import NDArray
from llama_stack.apis.files import Files
from llama_stack.apis.inference import InterleavedContent
from llama_stack.apis.inference import Inference, InterleavedContent
from llama_stack.apis.models import Models
from llama_stack.apis.vector_dbs import VectorDB
from llama_stack.apis.vector_io import (
Chunk,
QueryChunksResponse,
VectorIO,
)
from llama_stack.core.datatypes import VectorStoresConfig
from llama_stack.log import get_logger
from llama_stack.providers.datatypes import Api, VectorDBsProtocolPrivate
from llama_stack.providers.datatypes import VectorDBsProtocolPrivate
from llama_stack.providers.inline.vector_io.chroma import ChromaVectorIOConfig as InlineChromaVectorIOConfig
from llama_stack.providers.utils.kvstore import kvstore_impl
from llama_stack.providers.utils.kvstore.api import KVStore
@ -137,15 +139,17 @@ class ChromaVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
def __init__(
self,
config: RemoteChromaVectorIOConfig | InlineChromaVectorIOConfig,
inference_api: Api.inference,
models_apis: Api.models,
inference_api: Inference,
models_apis: Models,
files_api: Files | None,
vector_stores_config: VectorStoresConfig | None = None,
) -> None:
super().__init__(files_api=files_api, kvstore=None)
log.info(f"Initializing ChromaVectorIOAdapter with url: {config}")
self.config = config
self.inference_api = inference_api
self.models_api = models_apis
self.vector_stores_config = vector_stores_config
self.client = None
self.cache = {}
self.vector_db_store = None

11
llama_stack/providers/remote/vector_io/milvus/__init__.py
View file

@ -4,21 +4,28 @@
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
from llama_stack.core.datatypes import StackRunConfig
from llama_stack.providers.datatypes import Api, ProviderSpec
from .config import MilvusVectorIOConfig
async def get_adapter_impl(config: MilvusVectorIOConfig, deps: dict[Api, ProviderSpec]):
async def get_adapter_impl(
config: MilvusVectorIOConfig, deps: dict[Api, ProviderSpec], run_config: StackRunConfig | None = None
):
from .milvus import MilvusVectorIOAdapter
assert isinstance(config, MilvusVectorIOConfig), f"Unexpected config type: {type(config)}"
vector_stores_config = None
if run_config and run_config.vector_stores:
vector_stores_config = run_config.vector_stores
assert isinstance(config, MilvusVectorIOConfig), f"Unexpected config type: {type(config)}"
impl = MilvusVectorIOAdapter(
config,
deps[Api.inference],
deps[Api.models],
deps.get(Api.files),
vector_stores_config,
)
await impl.initialize()
return impl

5
llama_stack/providers/remote/vector_io/milvus/milvus.py
View file

@ -21,6 +21,7 @@ from llama_stack.apis.vector_io import (
QueryChunksResponse,
VectorIO,
)
from llama_stack.core.datatypes import VectorStoresConfig
from llama_stack.log import get_logger
from llama_stack.providers.datatypes import VectorDBsProtocolPrivate
from llama_stack.providers.inline.vector_io.milvus import MilvusVectorIOConfig as InlineMilvusVectorIOConfig
@ -308,8 +309,9 @@ class MilvusVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
self,
config: RemoteMilvusVectorIOConfig | InlineMilvusVectorIOConfig,
inference_api: Inference,
models_api: Models,
models_api: Models | None,
files_api: Files | None,
vector_stores_config: VectorStoresConfig | None = None,
) -> None:
super().__init__(files_api=files_api, kvstore=None)
self.config = config
@ -317,6 +319,7 @@ class MilvusVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
self.client = None
self.inference_api = inference_api
self.models_api = models_api
self.vector_stores_config = vector_stores_config
self.vector_db_store = None
self.metadata_collection_name = "openai_vector_stores_metadata"

16
llama_stack/providers/remote/vector_io/pgvector/__init__.py
View file

@ -4,14 +4,26 @@
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
from llama_stack.core.datatypes import StackRunConfig
from llama_stack.providers.datatypes import Api, ProviderSpec
from .config import PGVectorVectorIOConfig
async def get_adapter_impl(config: PGVectorVectorIOConfig, deps: dict[Api, ProviderSpec]):
async def get_adapter_impl(
config: PGVectorVectorIOConfig, deps: dict[Api, ProviderSpec], run_config: StackRunConfig | None = None
):
from .pgvector import PGVectorVectorIOAdapter
impl = PGVectorVectorIOAdapter(config, deps[Api.inference], deps[Api.models], deps.get(Api.files, None))
vector_stores_config = None
if run_config and run_config.vector_stores:
vector_stores_config = run_config.vector_stores
impl = PGVectorVectorIOAdapter(
config,
deps[Api.inference],
deps[Api.models],
deps.get(Api.files),
vector_stores_config,
)
await impl.initialize()
return impl

3
llama_stack/providers/remote/vector_io/pgvector/pgvector.py
View file

@ -23,6 +23,7 @@ from llama_stack.apis.vector_io import (
QueryChunksResponse,
VectorIO,
)
from llama_stack.core.datatypes import VectorStoresConfig
from llama_stack.log import get_logger
from llama_stack.providers.datatypes import VectorDBsProtocolPrivate
from llama_stack.providers.utils.inference.prompt_adapter import (
@ -346,11 +347,13 @@ class PGVectorVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtoco
inference_api: Inference,
models_api: Models,
files_api: Files | None = None,
vector_stores_config: VectorStoresConfig | None = None,
) -> None:
super().__init__(files_api=files_api, kvstore=None)
self.config = config
self.inference_api = inference_api
self.models_api = models_api
self.vector_stores_config = vector_stores_config
self.conn = None
self.cache = {}
self.vector_db_store = None

10
llama_stack/providers/remote/vector_io/qdrant/__init__.py
View file

@ -4,19 +4,27 @@
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
from llama_stack.core.datatypes import StackRunConfig
from llama_stack.providers.datatypes import Api, ProviderSpec
from .config import QdrantVectorIOConfig
async def get_adapter_impl(config: QdrantVectorIOConfig, deps: dict[Api, ProviderSpec]):
async def get_adapter_impl(
config: QdrantVectorIOConfig, deps: dict[Api, ProviderSpec], run_config: StackRunConfig | None = None
):
from .qdrant import QdrantVectorIOAdapter
vector_stores_config = None
if run_config and run_config.vector_stores:
vector_stores_config = run_config.vector_stores
impl = QdrantVectorIOAdapter(
config,
deps[Api.inference],
deps[Api.models],
deps.get(Api.files),
vector_stores_config,
)
await impl.initialize()
return impl

3
llama_stack/providers/remote/vector_io/qdrant/qdrant.py
View file

@ -25,6 +25,7 @@ from llama_stack.apis.vector_io import (
VectorStoreChunkingStrategy,
VectorStoreFileObject,
)
from llama_stack.core.datatypes import VectorStoresConfig
from llama_stack.log import get_logger
from llama_stack.providers.datatypes import VectorDBsProtocolPrivate
from llama_stack.providers.inline.vector_io.qdrant import QdrantVectorIOConfig as InlineQdrantVectorIOConfig
@ -163,6 +164,7 @@ class QdrantVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
inference_api: Inference,
models_api: Models,
files_api: Files | None = None,
vector_stores_config: VectorStoresConfig | None = None,
) -> None:
super().__init__(files_api=files_api, kvstore=None)
self.config = config
@ -170,6 +172,7 @@ class QdrantVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
self.cache = {}
self.inference_api = inference_api
self.models_api = models_api
self.vector_stores_config = vector_stores_config
self.vector_db_store = None
self._qdrant_lock = asyncio.Lock()

10
llama_stack/providers/remote/vector_io/weaviate/__init__.py
View file

@ -4,19 +4,27 @@
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
from llama_stack.core.datatypes import StackRunConfig
from llama_stack.providers.datatypes import Api, ProviderSpec
from .config import WeaviateVectorIOConfig
async def get_adapter_impl(config: WeaviateVectorIOConfig, deps: dict[Api, ProviderSpec]):
async def get_adapter_impl(
config: WeaviateVectorIOConfig, deps: dict[Api, ProviderSpec], run_config: StackRunConfig | None = None
):
from .weaviate import WeaviateVectorIOAdapter
vector_stores_config = None
if run_config and run_config.vector_stores:
vector_stores_config = run_config.vector_stores
impl = WeaviateVectorIOAdapter(
config,
deps[Api.inference],
deps[Api.models],
deps.get(Api.files),
vector_stores_config,
)
await impl.initialize()
return impl

3
llama_stack/providers/remote/vector_io/weaviate/weaviate.py
View file

@ -19,6 +19,7 @@ from llama_stack.apis.inference import Inference
from llama_stack.apis.models import Models
from llama_stack.apis.vector_dbs import VectorDB
from llama_stack.apis.vector_io import Chunk, QueryChunksResponse, VectorIO
from llama_stack.core.datatypes import VectorStoresConfig
from llama_stack.core.request_headers import NeedsRequestProviderData
from llama_stack.log import get_logger
from llama_stack.providers.datatypes import VectorDBsProtocolPrivate
@ -286,11 +287,13 @@ class WeaviateVectorIOAdapter(
inference_api: Inference,
models_api: Models,
files_api: Files | None,
vector_stores_config: VectorStoresConfig | None = None,
) -> None:
super().__init__(files_api=files_api, kvstore=None)
self.config = config
self.inference_api = inference_api
self.models_api = models_api
self.vector_stores_config = vector_stores_config
self.client_cache = {}
self.cache = {}
self.vector_db_store = None