mirror of
https://github.com/meta-llama/llama-stack.git
synced 2025-10-22 16:23:08 +00:00
chore: Updating how default embedding model is set in stack (#3818)
# What does this PR do?
Refactor setting default vector store provider and embedding model to
use an optional `vector_stores` config in the `StackRunConfig` and clean
up code to do so (had to add back in some pieces of VectorDB). Also
added remote Qdrant and Weaviate to starter distro (based on other PR
where inference providers were added for UX).
New config is simply (default for Starter distro):
```yaml
vector_stores:
default_provider_id: faiss
default_embedding_model:
provider_id: sentence-transformers
model_id: nomic-ai/nomic-embed-text-v1.5
```
## Test Plan
CI and Unit tests.
---------
Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>
Co-authored-by: Ashwin Bharambe <ashwin.bharambe@gmail.com>
This commit is contained in:
Francisco Arceo
GitHub
parent
2c43285e22
commit
48581bf651
48 changed files with 973 additions and 818 deletions
|
|
@ -25,6 +25,8 @@ distribution_spec:
|
|||
- provider_type: inline::milvus
|
||||
- provider_type: remote::chromadb
|
||||
- provider_type: remote::pgvector
|
||||
- provider_type: remote::qdrant
|
||||
- provider_type: remote::weaviate
|
||||
files:
|
||||
- provider_type: inline::localfs
|
||||
safety:
|
||||
|
|
|
|||
|
|
@ -128,6 +128,21 @@ providers:
|
|||
persistence:
|
||||
namespace: vector_io::pgvector
|
||||
backend: kv_default
|
||||
- provider_id: ${env.QDRANT_URL:+qdrant}
|
||||
provider_type: remote::qdrant
|
||||
config:
|
||||
api_key: ${env.QDRANT_API_KEY:=}
|
||||
persistence:
|
||||
namespace: vector_io::qdrant_remote
|
||||
backend: kv_default
|
||||
- provider_id: ${env.WEAVIATE_CLUSTER_URL:+weaviate}
|
||||
provider_type: remote::weaviate
|
||||
config:
|
||||
weaviate_api_key: null
|
||||
weaviate_cluster_url: ${env.WEAVIATE_CLUSTER_URL:=localhost:8080}
|
||||
persistence:
|
||||
namespace: vector_io::weaviate
|
||||
backend: kv_default
|
||||
files:
|
||||
- provider_id: meta-reference-files
|
||||
provider_type: inline::localfs
|
||||
|
|
@ -253,3 +268,8 @@ server:
|
|||
port: 8321
|
||||
telemetry:
|
||||
enabled: true
|
||||
vector_stores:
|
||||
default_provider_id: faiss
|
||||
default_embedding_model:
|
||||
provider_id: sentence-transformers
|
||||
model_id: nomic-ai/nomic-embed-text-v1.5
|
||||
|
|
|
|||
|
|
@ -26,6 +26,8 @@ distribution_spec:
|
|||
- provider_type: inline::milvus
|
||||
- provider_type: remote::chromadb
|
||||
- provider_type: remote::pgvector
|
||||
- provider_type: remote::qdrant
|
||||
- provider_type: remote::weaviate
|
||||
files:
|
||||
- provider_type: inline::localfs
|
||||
safety:
|
||||
|
|
|
|||
|
|
@ -128,6 +128,21 @@ providers:
|
|||
persistence:
|
||||
namespace: vector_io::pgvector
|
||||
backend: kv_default
|
||||
- provider_id: ${env.QDRANT_URL:+qdrant}
|
||||
provider_type: remote::qdrant
|
||||
config:
|
||||
api_key: ${env.QDRANT_API_KEY:=}
|
||||
persistence:
|
||||
namespace: vector_io::qdrant_remote
|
||||
backend: kv_default
|
||||
- provider_id: ${env.WEAVIATE_CLUSTER_URL:+weaviate}
|
||||
provider_type: remote::weaviate
|
||||
config:
|
||||
weaviate_api_key: null
|
||||
weaviate_cluster_url: ${env.WEAVIATE_CLUSTER_URL:=localhost:8080}
|
||||
persistence:
|
||||
namespace: vector_io::weaviate
|
||||
backend: kv_default
|
||||
files:
|
||||
- provider_id: meta-reference-files
|
||||
provider_type: inline::localfs
|
||||
|
|
@ -256,3 +271,8 @@ server:
|
|||
port: 8321
|
||||
telemetry:
|
||||
enabled: true
|
||||
vector_stores:
|
||||
default_provider_id: faiss
|
||||
default_embedding_model:
|
||||
provider_id: sentence-transformers
|
||||
model_id: nomic-ai/nomic-embed-text-v1.5
|
||||
|
|
|
|||
|
|
@ -26,6 +26,8 @@ distribution_spec:
|
|||
- provider_type: inline::milvus
|
||||
- provider_type: remote::chromadb
|
||||
- provider_type: remote::pgvector
|
||||
- provider_type: remote::qdrant
|
||||
- provider_type: remote::weaviate
|
||||
files:
|
||||
- provider_type: inline::localfs
|
||||
safety:
|
||||
|
|
|
|||
|
|
@ -128,6 +128,21 @@ providers:
|
|||
persistence:
|
||||
namespace: vector_io::pgvector
|
||||
backend: kv_default
|
||||
- provider_id: ${env.QDRANT_URL:+qdrant}
|
||||
provider_type: remote::qdrant
|
||||
config:
|
||||
api_key: ${env.QDRANT_API_KEY:=}
|
||||
persistence:
|
||||
namespace: vector_io::qdrant_remote
|
||||
backend: kv_default
|
||||
- provider_id: ${env.WEAVIATE_CLUSTER_URL:+weaviate}
|
||||
provider_type: remote::weaviate
|
||||
config:
|
||||
weaviate_api_key: null
|
||||
weaviate_cluster_url: ${env.WEAVIATE_CLUSTER_URL:=localhost:8080}
|
||||
persistence:
|
||||
namespace: vector_io::weaviate
|
||||
backend: kv_default
|
||||
files:
|
||||
- provider_id: meta-reference-files
|
||||
provider_type: inline::localfs
|
||||
|
|
@ -253,3 +268,8 @@ server:
|
|||
port: 8321
|
||||
telemetry:
|
||||
enabled: true
|
||||
vector_stores:
|
||||
default_provider_id: faiss
|
||||
default_embedding_model:
|
||||
provider_id: sentence-transformers
|
||||
model_id: nomic-ai/nomic-embed-text-v1.5
|
||||
|
|
|
|||
|
|
@ -11,8 +11,10 @@ from llama_stack.core.datatypes import (
|
|||
BuildProvider,
|
||||
Provider,
|
||||
ProviderSpec,
|
||||
QualifiedModel,
|
||||
ShieldInput,
|
||||
ToolGroupInput,
|
||||
VectorStoresConfig,
|
||||
)
|
||||
from llama_stack.core.utils.dynamic import instantiate_class_type
|
||||
from llama_stack.distributions.template import DistributionTemplate, RunConfigSettings
|
||||
|
|
@ -31,6 +33,8 @@ from llama_stack.providers.remote.vector_io.chroma.config import ChromaVectorIOC
|
|||
from llama_stack.providers.remote.vector_io.pgvector.config import (
|
||||
PGVectorVectorIOConfig,
|
||||
)
|
||||
from llama_stack.providers.remote.vector_io.qdrant.config import QdrantVectorIOConfig
|
||||
from llama_stack.providers.remote.vector_io.weaviate.config import WeaviateVectorIOConfig
|
||||
from llama_stack.providers.utils.sqlstore.sqlstore import PostgresSqlStoreConfig
|
||||
|
||||
|
||||
|
|
@ -113,6 +117,8 @@ def get_distribution_template(name: str = "starter") -> DistributionTemplate:
|
|||
BuildProvider(provider_type="inline::milvus"),
|
||||
BuildProvider(provider_type="remote::chromadb"),
|
||||
BuildProvider(provider_type="remote::pgvector"),
|
||||
BuildProvider(provider_type="remote::qdrant"),
|
||||
BuildProvider(provider_type="remote::weaviate"),
|
||||
],
|
||||
"files": [BuildProvider(provider_type="inline::localfs")],
|
||||
"safety": [
|
||||
|
|
@ -221,12 +227,35 @@ def get_distribution_template(name: str = "starter") -> DistributionTemplate:
|
|||
password="${env.PGVECTOR_PASSWORD:=}",
|
||||
),
|
||||
),
|
||||
Provider(
|
||||
provider_id="${env.QDRANT_URL:+qdrant}",
|
||||
provider_type="remote::qdrant",
|
||||
config=QdrantVectorIOConfig.sample_run_config(
|
||||
f"~/.llama/distributions/{name}",
|
||||
url="${env.QDRANT_URL:=}",
|
||||
),
|
||||
),
|
||||
Provider(
|
||||
provider_id="${env.WEAVIATE_CLUSTER_URL:+weaviate}",
|
||||
provider_type="remote::weaviate",
|
||||
config=WeaviateVectorIOConfig.sample_run_config(
|
||||
f"~/.llama/distributions/{name}",
|
||||
cluster_url="${env.WEAVIATE_CLUSTER_URL:=}",
|
||||
),
|
||||
),
|
||||
],
|
||||
"files": [files_provider],
|
||||
},
|
||||
default_models=[],
|
||||
default_tool_groups=default_tool_groups,
|
||||
default_shields=default_shields,
|
||||
vector_stores_config=VectorStoresConfig(
|
||||
default_provider_id="faiss",
|
||||
default_embedding_model=QualifiedModel(
|
||||
provider_id="sentence-transformers",
|
||||
model_id="nomic-ai/nomic-embed-text-v1.5",
|
||||
),
|
||||
),
|
||||
),
|
||||
},
|
||||
run_config_env_vars={
|
||||
|
|
|
|||
|
|
@ -27,6 +27,7 @@ from llama_stack.core.datatypes import (
|
|||
ShieldInput,
|
||||
TelemetryConfig,
|
||||
ToolGroupInput,
|
||||
VectorStoresConfig,
|
||||
)
|
||||
from llama_stack.core.distribution import get_provider_registry
|
||||
from llama_stack.core.storage.datatypes import (
|
||||
|
|
@ -186,6 +187,7 @@ class RunConfigSettings(BaseModel):
|
|||
default_tool_groups: list[ToolGroupInput] | None = None
|
||||
default_datasets: list[DatasetInput] | None = None
|
||||
default_benchmarks: list[BenchmarkInput] | None = None
|
||||
vector_stores_config: VectorStoresConfig | None = None
|
||||
telemetry: TelemetryConfig = Field(default_factory=lambda: TelemetryConfig(enabled=True))
|
||||
storage_backends: dict[str, Any] | None = None
|
||||
storage_stores: dict[str, Any] | None = None
|
||||
|
|
@ -263,7 +265,7 @@ class RunConfigSettings(BaseModel):
|
|||
)
|
||||
|
||||
# Return a dict that matches StackRunConfig structure
|
||||
return {
|
||||
config = {
|
||||
"version": LLAMA_STACK_RUN_CONFIG_VERSION,
|
||||
"image_name": name,
|
||||
"container_image": container_image,
|
||||
|
|
@ -283,6 +285,11 @@ class RunConfigSettings(BaseModel):
|
|||
"telemetry": self.telemetry.model_dump(exclude_none=True) if self.telemetry else None,
|
||||
}
|
||||
|
||||
if self.vector_stores_config:
|
||||
config["vector_stores"] = self.vector_stores_config.model_dump(exclude_none=True)
|
||||
|
||||
return config
|
||||
|
||||
|
||||
class DistributionTemplate(BaseModel):
|
||||
"""
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue