phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-10-23 00:27:26 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

chore(cleanup)!: kill vector_db references as far as possible (#3864)

Browse source 
There should not be "vector db" anywhere.
This commit is contained in:
Ashwin Bharambe 2025-10-20 20:06:16 -07:00 • committed by GitHub
parent 444f6c88f3
commit 122de785c4
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
46 changed files with 701 additions and 822 deletions
Download patch file Download diff file Expand all files Collapse all files

2
llama_stack/apis/datatypes.py
View file

@ -121,7 +121,7 @@ class Api(Enum, metaclass=DynamicApiMeta):
models = "models"
shields = "shields"
vector_dbs = "vector_dbs" # only used for routing
vector_stores = "vector_stores" # only used for routing table
datasets = "datasets"
scoring_functions = "scoring_functions"
benchmarks = "benchmarks"

4
llama_stack/apis/resource.py
View file

@ -13,7 +13,7 @@ from pydantic import BaseModel, Field
class ResourceType(StrEnum):
model = "model"
shield = "shield"
vector_db = "vector_db"
vector_store = "vector_store"
dataset = "dataset"
scoring_function = "scoring_function"
benchmark = "benchmark"
@ -34,4 +34,4 @@ class Resource(BaseModel):
provider_id: str = Field(description="ID of the provider that owns this resource")
type: ResourceType = Field(description="Type of resource (e.g. 'model', 'shield', 'vector_db', etc.)")
type: ResourceType = Field(description="Type of resource (e.g. 'model', 'shield', 'vector_store', etc.)")

93
llama_stack/apis/vector_dbs/vector_dbs.py
View file

@ -1,93 +0,0 @@
# 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 Literal, Protocol, runtime_checkable
from pydantic import BaseModel
from llama_stack.apis.resource import Resource, ResourceType
from llama_stack.schema_utils import json_schema_type
@json_schema_type
class VectorDB(Resource):
"""Vector database resource for storing and querying vector embeddings.
:param type: Type of resource, always 'vector_db' for vector databases
:param embedding_model: Name of the embedding model to use for vector generation
:param embedding_dimension: Dimension of the embedding vectors
"""
type: Literal[ResourceType.vector_db] = ResourceType.vector_db
embedding_model: str
embedding_dimension: int
vector_db_name: str | None = None
@property
def vector_db_id(self) -> str:
return self.identifier
@property
def provider_vector_db_id(self) -> str | None:
return self.provider_resource_id
class VectorDBInput(BaseModel):
"""Input parameters for creating or configuring a vector database.
:param vector_db_id: Unique identifier for the vector database
:param embedding_model: Name of the embedding model to use for vector generation
:param embedding_dimension: Dimension of the embedding vectors
:param provider_vector_db_id: (Optional) Provider-specific identifier for the vector database
"""
vector_db_id: str
embedding_model: str
embedding_dimension: int
provider_id: str | None = None
provider_vector_db_id: str | None = None
class ListVectorDBsResponse(BaseModel):
"""Response from listing vector databases.
:param data: List of vector databases
"""
data: list[VectorDB]
@runtime_checkable
class VectorDBs(Protocol):
"""Internal protocol for vector_dbs routing - no public API endpoints."""
async def list_vector_dbs(self) -> ListVectorDBsResponse:
"""Internal method to list vector databases."""
...
async def get_vector_db(
self,
vector_db_id: str,
) -> VectorDB:
"""Internal method to get a vector database by ID."""
...
async def register_vector_db(
self,
vector_db_id: str,
embedding_model: str,
embedding_dimension: int | None = 384,
provider_id: str | None = None,
vector_db_name: str | None = None,
provider_vector_db_id: str | None = None,
) -> VectorDB:
"""Internal method to register a vector database."""
...
async def unregister_vector_db(self, vector_db_id: str) -> None:
"""Internal method to unregister a vector database."""
...

11
llama_stack/apis/vector_io/vector_io.py
View file

@ -15,7 +15,7 @@ from fastapi import Body
from pydantic import BaseModel, Field
from llama_stack.apis.inference import InterleavedContent
from llama_stack.apis.vector_dbs import VectorDB
from llama_stack.apis.vector_stores import VectorStore
from llama_stack.apis.version import LLAMA_STACK_API_V1
from llama_stack.providers.utils.telemetry.trace_protocol import trace_protocol
from llama_stack.providers.utils.vector_io.vector_utils import generate_chunk_id
@ -140,6 +140,7 @@ class VectorStoreFileCounts(BaseModel):
total: int
# TODO: rename this as OpenAIVectorStore
@json_schema_type
class VectorStoreObject(BaseModel):
"""OpenAI Vector Store object.
@ -517,17 +518,18 @@ class OpenAICreateVectorStoreFileBatchRequestWithExtraBody(BaseModel, extra="all
chunking_strategy: VectorStoreChunkingStrategy | None = None
class VectorDBStore(Protocol):
def get_vector_db(self, vector_db_id: str) -> VectorDB | None: ...
class VectorStoreTable(Protocol):
def get_vector_store(self, vector_store_id: str) -> VectorStore | None: ...
@runtime_checkable
@trace_protocol
class VectorIO(Protocol):
vector_db_store: VectorDBStore | None = None
vector_store_table: VectorStoreTable | None = None
# this will just block now until chunks are inserted, but it should
# probably return a Job instance which can be polled for completion
# TODO: rename vector_db_id to vector_store_id once Stainless is working
@webmethod(route="/vector-io/insert", method="POST", level=LLAMA_STACK_API_V1)
async def insert_chunks(
self,
@ -546,6 +548,7 @@ class VectorIO(Protocol):
"""
...
# TODO: rename vector_db_id to vector_store_id once Stainless is working
@webmethod(route="/vector-io/query", method="POST", level=LLAMA_STACK_API_V1)
async def query_chunks(
self,

2
llama_stack/apis/vector_dbs/__init__.py → llama_stack/apis/vector_stores/__init__.py
View file

@ -4,4 +4,4 @@
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
from .vector_dbs import *
from .vector_stores import *

51
llama_stack/apis/vector_stores/vector_stores.py Normal file
View file

@ -0,0 +1,51 @@
# 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 Literal
from pydantic import BaseModel
from llama_stack.apis.resource import Resource, ResourceType
# Internal resource type for storing the vector store routing and other information
class VectorStore(Resource):
"""Vector database resource for storing and querying vector embeddings.
:param type: Type of resource, always 'vector_store' for vector stores
:param embedding_model: Name of the embedding model to use for vector generation
:param embedding_dimension: Dimension of the embedding vectors
"""
type: Literal[ResourceType.vector_store] = ResourceType.vector_store
embedding_model: str
embedding_dimension: int
vector_store_name: str | None = None
@property
def vector_store_id(self) -> str:
return self.identifier
@property
def provider_vector_store_id(self) -> str | None:
return self.provider_resource_id
class VectorStoreInput(BaseModel):
"""Input parameters for creating or configuring a vector database.
:param vector_store_id: Unique identifier for the vector store
:param embedding_model: Name of the embedding model to use for vector generation
:param embedding_dimension: Dimension of the embedding vectors
:param provider_vector_store_id: (Optional) Provider-specific identifier for the vector store
"""
vector_store_id: str
embedding_model: str
embedding_dimension: int
provider_id: str | None = None
provider_vector_store_id: str | None = None