phoenix-oss/llama-stack
1
0
Fork 0
forked from phoenix-oss/llama-stack-mirror
Code Pull requests Releases Packages 2 Activity Actions

[memory refactor][1/n] Rename Memory -> VectorIO, MemoryBanks -> VectorDBs (#828)

Browse source 
See https://github.com/meta-llama/llama-stack/issues/827 for the broader
design.

This is the first part:

- delete other kinds of memory banks (keyvalue, keyword, graph) for now;
we will introduce a keyvalue store API as part of this design but not
use it in the RAG tool yet.
- renaming of the APIs
This commit is contained in:
Ashwin Bharambe 2025-01-22 09:59:30 -08:00 • committed by GitHub
parent 35a00d004a
commit 3ae8585b65
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
37 changed files with 175 additions and 296 deletions
Download patch file Download diff file Expand all files Collapse all files

50
llama_stack/providers/registry/memory.py → llama_stack/providers/registry/vector_io.py
View file

@ -38,78 +38,78 @@ EMBEDDING_DEPS = [
def available_providers() -> List[ProviderSpec]:
return [
InlineProviderSpec(
api=Api.memory,
api=Api.vector_io,
provider_type="inline::meta-reference",
pip_packages=EMBEDDING_DEPS + ["faiss-cpu"],
module="llama_stack.providers.inline.memory.faiss",
config_class="llama_stack.providers.inline.memory.faiss.FaissImplConfig",
module="llama_stack.providers.inline.vector_io.faiss",
config_class="llama_stack.providers.inline.vector_io.faiss.FaissImplConfig",
deprecation_warning="Please use the `inline::faiss` provider instead.",
api_dependencies=[Api.inference],
),
InlineProviderSpec(
api=Api.memory,
api=Api.vector_io,
provider_type="inline::faiss",
pip_packages=EMBEDDING_DEPS + ["faiss-cpu"],
module="llama_stack.providers.inline.memory.faiss",
config_class="llama_stack.providers.inline.memory.faiss.FaissImplConfig",
module="llama_stack.providers.inline.vector_io.faiss",
config_class="llama_stack.providers.inline.vector_io.faiss.FaissImplConfig",
api_dependencies=[Api.inference],
),
remote_provider_spec(
Api.memory,
Api.vector_io,
AdapterSpec(
adapter_type="chromadb",
pip_packages=EMBEDDING_DEPS + ["chromadb-client"],
module="llama_stack.providers.remote.memory.chroma",
config_class="llama_stack.providers.remote.memory.chroma.ChromaRemoteImplConfig",
module="llama_stack.providers.remote.vector_io.chroma",
config_class="llama_stack.providers.remote.vector_io.chroma.ChromaRemoteImplConfig",
),
api_dependencies=[Api.inference],
),
InlineProviderSpec(
api=Api.memory,
api=Api.vector_io,
provider_type="inline::chromadb",
pip_packages=EMBEDDING_DEPS + ["chromadb"],
module="llama_stack.providers.inline.memory.chroma",
config_class="llama_stack.providers.inline.memory.chroma.ChromaInlineImplConfig",
module="llama_stack.providers.inline.vector_io.chroma",
config_class="llama_stack.providers.inline.vector_io.chroma.ChromaInlineImplConfig",
api_dependencies=[Api.inference],
),
remote_provider_spec(
Api.memory,
Api.vector_io,
AdapterSpec(
adapter_type="pgvector",
pip_packages=EMBEDDING_DEPS + ["psycopg2-binary"],
module="llama_stack.providers.remote.memory.pgvector",
config_class="llama_stack.providers.remote.memory.pgvector.PGVectorConfig",
module="llama_stack.providers.remote.vector_io.pgvector",
config_class="llama_stack.providers.remote.vector_io.pgvector.PGVectorConfig",
),
api_dependencies=[Api.inference],
),
remote_provider_spec(
Api.memory,
Api.vector_io,
AdapterSpec(
adapter_type="weaviate",
pip_packages=EMBEDDING_DEPS + ["weaviate-client"],
module="llama_stack.providers.remote.memory.weaviate",
config_class="llama_stack.providers.remote.memory.weaviate.WeaviateConfig",
provider_data_validator="llama_stack.providers.remote.memory.weaviate.WeaviateRequestProviderData",
module="llama_stack.providers.remote.vector_io.weaviate",
config_class="llama_stack.providers.remote.vector_io.weaviate.WeaviateConfig",
provider_data_validator="llama_stack.providers.remote.vector_io.weaviate.WeaviateRequestProviderData",
),
api_dependencies=[Api.inference],
),
remote_provider_spec(
api=Api.memory,
api=Api.vector_io,
adapter=AdapterSpec(
adapter_type="sample",
pip_packages=[],
module="llama_stack.providers.remote.memory.sample",
config_class="llama_stack.providers.remote.memory.sample.SampleConfig",
module="llama_stack.providers.remote.vector_io.sample",
config_class="llama_stack.providers.remote.vector_io.sample.SampleConfig",
),
api_dependencies=[],
),
remote_provider_spec(
Api.memory,
Api.vector_io,
AdapterSpec(
adapter_type="qdrant",
pip_packages=EMBEDDING_DEPS + ["qdrant-client"],
module="llama_stack.providers.remote.memory.qdrant",
config_class="llama_stack.providers.remote.memory.qdrant.QdrantConfig",
module="llama_stack.providers.remote.vector_io.qdrant",
config_class="llama_stack.providers.remote.vector_io.qdrant.QdrantConfig",
),
api_dependencies=[Api.inference],
),