phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-07-15 17:44:01 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'main' into chunk-metadata

Browse source
This commit is contained in:
Francisco Arceo 2025-06-25 12:57:50 -06:00 • committed by GitHub
commit f52eb51555
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
40 changed files with 272 additions and 722 deletions
Download patch file Download diff file Expand all files Collapse all files

25
llama_stack/providers/utils/memory/openai_vector_store_mixin.py
View file

@ -16,20 +16,14 @@ from llama_stack.apis.files import Files
from llama_stack.apis.files.files import OpenAIFileObject
from llama_stack.apis.vector_dbs import VectorDB
from llama_stack.apis.vector_io import (
Chunk,
QueryChunksResponse,
SearchRankingOptions,
VectorStoreContent,
VectorStoreDeleteResponse,
VectorStoreListResponse,
VectorStoreObject,
VectorStoreSearchResponse,
VectorStoreSearchResponsePage,
)
from llama_stack.apis.vector_io.vector_io import (
Chunk,
VectorStoreChunkingStrategy,
VectorStoreChunkingStrategyAuto,
VectorStoreChunkingStrategyStatic,
VectorStoreContent,
VectorStoreDeleteResponse,
VectorStoreFileContentsResponse,
VectorStoreFileCounts,
VectorStoreFileDeleteResponse,
@ -37,6 +31,10 @@ from llama_stack.apis.vector_io.vector_io import (
VectorStoreFileObject,
VectorStoreFileStatus,
VectorStoreListFilesResponse,
VectorStoreListResponse,
VectorStoreObject,
VectorStoreSearchResponse,
VectorStoreSearchResponsePage,
)
from llama_stack.providers.utils.memory.vector_store import content_from_data_and_mime_type, make_overlapped_chunks
@ -339,13 +337,16 @@ class OpenAIVectorStoreMixin(ABC):
max_num_results: int | None = 10,
ranking_options: SearchRankingOptions | None = None,
rewrite_query: bool | None = False,
# search_mode: Literal["keyword", "vector", "hybrid"] = "vector",
search_mode: str | None = "vector", # Using str instead of Literal due to OpenAPI schema generator limitations
) -> VectorStoreSearchResponsePage:
"""Search for chunks in a vector store."""
# TODO: Add support in the API for this
search_mode = "vector"
max_num_results = max_num_results or 10
# Validate search_mode
valid_modes = {"keyword", "vector", "hybrid"}
if search_mode not in valid_modes:
raise ValueError(f"search_mode must be one of {valid_modes}, got {search_mode}")
if vector_store_id not in self.openai_vector_stores:
raise ValueError(f"Vector store {vector_store_id} not found")