phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-22 22:42:25 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'main' into pgvector-openai-compat

Browse source
This commit is contained in:
Francisco Arceo 2025-07-15 11:36:57 -04:00 • committed by GitHub
commit c0022be33d
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
11 changed files with 95 additions and 103 deletions
Download patch file Download diff file Expand all files Collapse all files

69
.github/workflows/tests.yml vendored
View file

@ -1,69 +0,0 @@
name: auto-tests
on:
# pull_request:
workflow_dispatch:
inputs:
commit_sha:
description: 'Specific Commit SHA to trigger on'
required: false
default: $GITHUB_SHA # default to the last commit of $GITHUB_REF branch
jobs:
test-llama-stack-as-library:
runs-on: ubuntu-latest
env:
TOGETHER_API_KEY: ${{ secrets.TOGETHER_API_KEY }}
FIREWORKS_API_KEY: ${{ secrets.FIREWORKS_API_KEY }}
TAVILY_SEARCH_API_KEY: ${{ secrets.TAVILY_SEARCH_API_KEY }}
strategy:
matrix:
provider: [fireworks, together]
steps:
- uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
with:
ref: ${{ github.event.inputs.commit_sha }}
- name: Echo commit SHA
run: |
echo "Triggered on commit SHA: ${{ github.event.inputs.commit_sha }}"
git rev-parse HEAD
- name: Install dependencies
run: |
python -m pip install --upgrade pip
pip install -r requirements.txt pytest
pip install -e .
- name: Build providers
run: |
llama stack build --template ${{ matrix.provider }} --image-type venv
- name: Install the latest llama-stack-client & llama-models packages
run: |
pip install -e git+https://github.com/meta-llama/llama-stack-client-python.git#egg=llama-stack-client
pip install -e git+https://github.com/meta-llama/llama-models.git#egg=llama-models
- name: Run client-sdk test
working-directory: "${{ github.workspace }}"
env:
REPORT_OUTPUT: md_report.md
shell: bash
run: |
pip install --upgrade pytest-md-report
echo "REPORT_FILE=${REPORT_OUTPUT}" >> "$GITHUB_ENV"
export INFERENCE_MODEL=meta-llama/Llama-3.1-8B-Instruct
LLAMA_STACK_CONFIG=./llama_stack/templates/${{ matrix.provider }}/run.yaml pytest --md-report --md-report-verbose=1 ./tests/client-sdk/inference/ --md-report-output "$REPORT_OUTPUT"
- name: Output reports to the job summary
if: always()
shell: bash
run: |
if [ -f "$REPORT_FILE" ]; then
echo "<details><summary> Test Report for ${{ matrix.provider }} </summary>" >> $GITHUB_STEP_SUMMARY
echo "" >> $GITHUB_STEP_SUMMARY
cat "$REPORT_FILE" >> $GITHUB_STEP_SUMMARY
echo "" >> $GITHUB_STEP_SUMMARY
echo "</details>" >> $GITHUB_STEP_SUMMARY
fi

11
docs/_static/llama-stack-spec.html vendored
View file

@ -11340,6 +11340,9 @@
},
"embedding_dimension": {
"type": "integer"
},
"vector_db_name": {
"type": "string"
}
},
"additionalProperties": false,
@ -13590,10 +13593,6 @@
"provider_id": {
"type": "string",
"description": "The ID of the provider to use for this vector store."
},
"provider_vector_db_id": {
"type": "string",
"description": "The provider-specific vector database ID."
}
},
"additionalProperties": false,
@ -15634,6 +15633,10 @@
"type": "string",
"description": "The identifier of the provider."
},
"vector_db_name": {
"type": "string",
"description": "The name of the vector database."
},
"provider_vector_db_id": {
"type": "string",
"description": "The identifier of the vector database in the provider."

9
docs/_static/llama-stack-spec.yaml vendored
View file

@ -7984,6 +7984,8 @@ components:
type: string
embedding_dimension:
type: integer
vector_db_name:
type: string
additionalProperties: false
required:
- identifier
@ -9494,10 +9496,6 @@ components:
type: string
description: >-
The ID of the provider to use for this vector store.
provider_vector_db_id:
type: string
description: >-
The provider-specific vector database ID.
additionalProperties: false
required:
- name
@ -10945,6 +10943,9 @@ components:
provider_id:
type: string
description: The identifier of the provider.
vector_db_name:
type: string
description: The name of the vector database.
provider_vector_db_id:
type: string
description: >-

3
llama_stack/apis/vector_dbs/vector_dbs.py
View file

@ -19,6 +19,7 @@ class VectorDB(Resource):
embedding_model: str
embedding_dimension: int
vector_db_name: str | None = None
@property
def vector_db_id(self) -> str:
@ -70,6 +71,7 @@ class VectorDBs(Protocol):
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:
"""Register a vector database.
@ -78,6 +80,7 @@ class VectorDBs(Protocol):
:param embedding_model: The embedding model to use.
:param embedding_dimension: The dimension of the embedding model.
:param provider_id: The identifier of the provider.
:param vector_db_name: The name of the vector database.
:param provider_vector_db_id: The identifier of the vector database in the provider.
:returns: A VectorDB.
"""

2
llama_stack/apis/vector_io/vector_io.py
View file

@ -346,7 +346,6 @@ class VectorIO(Protocol):
embedding_model: str | None = None,
embedding_dimension: int | None = 384,
provider_id: str | None = None,
provider_vector_db_id: str | None = None,
) -> VectorStoreObject:
"""Creates a vector store.
@ -358,7 +357,6 @@ class VectorIO(Protocol):
:param embedding_model: The embedding model to use for this vector store.
:param embedding_dimension: The dimension of the embedding vectors (default: 384).
:param provider_id: The ID of the provider to use for this vector store.
:param provider_vector_db_id: The provider-specific vector database ID.
:returns: A VectorStoreObject representing the created vector store.
"""
...

20
llama_stack/distribution/routers/vector_io.py
View file

@ -5,6 +5,7 @@
# the root directory of this source tree.
import asyncio
import uuid
from typing import Any
from llama_stack.apis.common.content_types import (
@ -81,6 +82,7 @@ class VectorIORouter(VectorIO):
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,
) -> None:
logger.debug(f"VectorIORouter.register_vector_db: {vector_db_id}, {embedding_model}")
@ -89,6 +91,7 @@ class VectorIORouter(VectorIO):
embedding_model,
embedding_dimension,
provider_id,
vector_db_name,
provider_vector_db_id,
)
@ -123,7 +126,6 @@ class VectorIORouter(VectorIO):
embedding_model: str | None = None,
embedding_dimension: int | None = None,
provider_id: str | None = None,
provider_vector_db_id: str | None = None,
) -> VectorStoreObject:
logger.debug(f"VectorIORouter.openai_create_vector_store: name={name}, provider_id={provider_id}")
@ -135,17 +137,17 @@ class VectorIORouter(VectorIO):
embedding_model, embedding_dimension = embedding_model_info
logger.info(f"No embedding model specified, using first available: {embedding_model}")
vector_db_id = name
vector_db_id = f"vs_{uuid.uuid4()}"
registered_vector_db = await self.routing_table.register_vector_db(
vector_db_id,
embedding_model,
embedding_dimension,
provider_id,
provider_vector_db_id,
vector_db_id=vector_db_id,
embedding_model=embedding_model,
embedding_dimension=embedding_dimension,
provider_id=provider_id,
provider_vector_db_id=vector_db_id,
vector_db_name=name,
)
return await self.routing_table.get_provider_impl(registered_vector_db.identifier).openai_create_vector_store(
vector_db_id,
name=name,
file_ids=file_ids,
expires_after=expires_after,
chunking_strategy=chunking_strategy,

2
llama_stack/distribution/routing_tables/vector_dbs.py
View file

@ -36,6 +36,7 @@ class VectorDBsRoutingTable(CommonRoutingTableImpl, VectorDBs):
embedding_dimension: int | None = 384,
provider_id: str | None = None,
provider_vector_db_id: str | None = None,
vector_db_name: str | None = None,
) -> VectorDB:
if provider_vector_db_id is None:
provider_vector_db_id = vector_db_id
@ -62,6 +63,7 @@ class VectorDBsRoutingTable(CommonRoutingTableImpl, VectorDBs):
"provider_resource_id": provider_vector_db_id,
"embedding_model": embedding_model,
"embedding_dimension": model.metadata["embedding_dimension"],
"vector_db_name": vector_db_name,
}
vector_db = TypeAdapter(VectorDBWithOwner).validate_python(vector_db_data)
await self.register_object(vector_db)

1
llama_stack/providers/remote/vector_io/chroma/chroma.py
View file

@ -217,7 +217,6 @@ class ChromaVectorIOAdapter(VectorIO, VectorDBsProtocolPrivate):
embedding_model: str | None = None,
embedding_dimension: int | None = 384,
provider_id: str | None = None,
provider_vector_db_id: str | None = None,
) -> VectorStoreObject:
raise NotImplementedError("OpenAI Vector Stores API is not supported in Chroma")

1
llama_stack/providers/remote/vector_io/qdrant/qdrant.py
View file

@ -214,7 +214,6 @@ class QdrantVectorIOAdapter(VectorIO, VectorDBsProtocolPrivate):
embedding_model: str | None = None,
embedding_dimension: int | None = 384,
provider_id: str | None = None,
provider_vector_db_id: str | None = None,
) -> VectorStoreObject:
raise NotImplementedError("OpenAI Vector Stores API is not supported in Qdrant")

27
llama_stack/providers/utils/memory/openai_vector_store_mixin.py
View file

@ -172,8 +172,9 @@ class OpenAIVectorStoreMixin(ABC):
provider_vector_db_id: str | None = None,
) -> VectorStoreObject:
"""Creates a vector store."""
store_id = name or str(uuid.uuid4())
created_at = int(time.time())
# Derive the canonical vector_db_id (allow override, else generate)
vector_db_id = provider_vector_db_id or f"vs_{uuid.uuid4()}"
if provider_id is None:
raise ValueError("Provider ID is required")
@ -181,19 +182,19 @@ class OpenAIVectorStoreMixin(ABC):
if embedding_model is None:
raise ValueError("Embedding model is required")
# Use provided embedding dimension or default to 384
# Embedding dimension is required (defaulted to 384 if not provided)
if embedding_dimension is None:
raise ValueError("Embedding dimension is required")
provider_vector_db_id = provider_vector_db_id or store_id
# Register the VectorDB backing this vector store
vector_db = VectorDB(
identifier=store_id,
identifier=vector_db_id,
embedding_dimension=embedding_dimension,
embedding_model=embedding_model,
provider_id=provider_id,
provider_resource_id=provider_vector_db_id,
provider_resource_id=vector_db_id,
vector_db_name=name,
)
# Register the vector DB
await self.register_vector_db(vector_db)
# Create OpenAI vector store metadata
@ -207,11 +208,11 @@ class OpenAIVectorStoreMixin(ABC):
in_progress=0,
total=0,
)
store_info = {
"id": store_id,
store_info: dict[str, Any] = {
"id": vector_db_id,
"object": "vector_store",
"created_at": created_at,
"name": store_id,
"name": name,
"usage_bytes": 0,
"file_counts": file_counts.model_dump(),
"status": status,
@ -231,18 +232,18 @@ class OpenAIVectorStoreMixin(ABC):
store_info["metadata"] = metadata
# Save to persistent storage (provider-specific)
await self._save_openai_vector_store(store_id, store_info)
await self._save_openai_vector_store(vector_db_id, store_info)
# Store in memory cache
self.openai_vector_stores[store_id] = store_info
self.openai_vector_stores[vector_db_id] = store_info
# Now that our vector store is created, attach any files that were provided
file_ids = file_ids or []
tasks = [self.openai_attach_file_to_vector_store(store_id, file_id) for file_id in file_ids]
tasks = [self.openai_attach_file_to_vector_store(vector_db_id, file_id) for file_id in file_ids]
await asyncio.gather(*tasks)
# Get the updated store info and return it
store_info = self.openai_vector_stores[store_id]
store_info = self.openai_vector_stores[vector_db_id]
return VectorStoreObject.model_validate(store_info)
async def openai_list_vector_stores(

53
tests/integration/vector_io/test_openai_vector_stores.py
View file

@ -821,6 +821,59 @@ def test_openai_vector_store_update_file(compat_client_with_empty_stores, client
assert retrieved_file.attributes["foo"] == "baz"
def test_create_vector_store_files_duplicate_vector_store_name(compat_client_with_empty_stores, client_with_models):
"""
This test confirms that client.vector_stores.create() creates a unique ID
"""
skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
skip_if_provider_doesnt_support_openai_vector_store_files_api(client_with_models)
if isinstance(compat_client_with_empty_stores, LlamaStackClient):
pytest.skip("Vector Store Files create is not yet supported with LlamaStackClient")
compat_client = compat_client_with_empty_stores
# Create a vector store with files
file_ids = []
for i in range(3):
with BytesIO(f"This is a test file {i}".encode()) as file_buffer:
file_buffer.name = f"openai_test_{i}.txt"
file = compat_client.files.create(file=file_buffer, purpose="assistants")
file_ids.append(file.id)
vector_store = compat_client.vector_stores.create(
name="test_store_with_files",
)
assert vector_store.file_counts.completed == 0
assert vector_store.file_counts.total == 0
assert vector_store.file_counts.cancelled == 0
assert vector_store.file_counts.failed == 0
assert vector_store.file_counts.in_progress == 0
vector_store2 = compat_client.vector_stores.create(
name="test_store_with_files",
)
vector_stores_list = compat_client.vector_stores.list()
assert len(vector_stores_list.data) == 2
created_file = compat_client.vector_stores.files.create(
vector_store_id=vector_store.id,
file_id=file_ids[0],
)
assert created_file.status == "completed"
_ = compat_client.vector_stores.delete(vector_store2.id)
created_file_from_non_deleted_vector_store = compat_client.vector_stores.files.create(
vector_store_id=vector_store.id,
file_id=file_ids[1],
)
assert created_file_from_non_deleted_vector_store.status == "completed"
vector_stores_list_post_delete = compat_client.vector_stores.list()
assert len(vector_stores_list_post_delete.data) == 1
@pytest.mark.skip(reason="Client library needs to be scaffolded to support search_mode parameter")
def test_openai_vector_store_search_modes():
"""Test OpenAI vector store search with different search modes.