phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-15 12:02:37 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

update resolver to only pass vector_stores section of run config

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

Using Router only from VectorDBs

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

removing model_api from vector store providers

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

fix test

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

updating integration tests

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

special handling for replay mode for available providers

Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>
This commit is contained in:
Francisco Javier Arceo 2025-10-16 10:59:01 -04:00
parent 24a1430c8b
commit accc4c437e
46 changed files with 397 additions and 702 deletions
Download patch file Download diff file Expand all files Collapse all files

3
llama_stack/core/routers/__init__.py
View file

@ -84,6 +84,9 @@ 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:
api_to_dep_impl["vector_stores_config"] = run_config.vector_stores
impl = api_to_routers[api.value](routing_table, **api_to_dep_impl)
await impl.initialize()
return impl

7
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
@ -43,9 +44,11 @@ class VectorIORouter(VectorIO):
def __init__(
self,
routing_table: RoutingTable,
vector_stores_config: VectorStoresConfig | None = None,
) -> None:
logger.debug("Initializing VectorIORouter")
self.routing_table = routing_table
self.vector_stores_config = vector_stores_config
async def initialize(self) -> None:
logger.debug("VectorIORouter.initialize")
@ -122,6 +125,10 @@ class VectorIORouter(VectorIO):
embedding_dimension = extra.get("embedding_dimension")
provider_id = extra.get("provider_id")
if embedding_model is None and self.vector_stores_config is not None:
embedding_model = self.vector_stores_config.default_embedding_model_id
logger.debug(f"Using default embedding model: {embedding_model}")
if embedding_model is not None and embedding_dimension is None:
embedding_dimension = await self._get_embedding_model_dimension(embedding_model)