phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-12 04:00:42 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

more small cleanup

Browse source
This commit is contained in:
Ashwin Bharambe 2025-10-20 14:09:44 -07:00
parent 00c6bbffb7
commit 9c9f5f059a
2 changed files with 12 additions and 8 deletions
Download patch file Download diff file Expand all files Collapse all files

2
llama_stack/core/routers/__init__.py
View file

@ -93,7 +93,7 @@ async def get_auto_router_impl(
await inference_store.initialize()
api_to_dep_impl["store"] = inference_store
if api == Api.vector_io and run_config.vector_stores:
elif api == Api.vector_io:
api_to_dep_impl["vector_stores_config"] = run_config.vector_stores
impl = api_to_routers[api.value](routing_table, **api_to_dep_impl)

18
llama_stack/core/routers/vector_io.py
View file

@ -31,6 +31,7 @@ from llama_stack.apis.vector_io import (
VectorStoreObject,
VectorStoreSearchResponsePage,
)
from llama_stack.core.datatypes import VectorStoresConfig
from llama_stack.log import get_logger
from llama_stack.providers.datatypes import HealthResponse, HealthStatus, RoutingTable
@ -43,7 +44,7 @@ class VectorIORouter(VectorIO):
def __init__(
self,
routing_table: RoutingTable,
vector_stores_config=None,
vector_stores_config: VectorStoresConfig | None = None,
) -> None:
logger.debug("Initializing VectorIORouter")
self.routing_table = routing_table
@ -125,12 +126,15 @@ class VectorIORouter(VectorIO):
provider_id = extra.get("provider_id")
# Use default embedding model if not specified
if embedding_model is None and self.vector_stores_config is not None:
if self.vector_stores_config.default_embedding_model is not None:
# Construct the full model ID with provider prefix
embedding_provider_id = self.vector_stores_config.default_embedding_model.provider_id
model_id = self.vector_stores_config.default_embedding_model.model_id
embedding_model = f"{embedding_provider_id}/{model_id}"
if (
embedding_model is None
and self.vector_stores_config
and self.vector_stores_config.default_embedding_model is not None
):
# Construct the full model ID with provider prefix
embedding_provider_id = self.vector_stores_config.default_embedding_model.provider_id
model_id = self.vector_stores_config.default_embedding_model.model_id
embedding_model = f"{embedding_provider_id}/{model_id}"
if embedding_model is not None and embedding_dimension is None:
embedding_dimension = await self._get_embedding_model_dimension(embedding_model)