phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-07-16 18:08:09 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

feat: Enable ingestion of custom embeddings

Browse source 
Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>
This commit is contained in:
Francisco Javier Arceo 2025-05-29 20:58:41 -04:00 • committed by Francisco Arceo
parent 2603f10f95
commit 73456878e5
8 changed files with 224 additions and 15 deletions
Download patch file Download diff file Expand all files Collapse all files

34
tests/integration/vector_io/test_vector_io.py
View file

@ -120,3 +120,37 @@ def test_insert_chunks(client_with_empty_registry, embedding_model_id, sample_ch
top_match = response.chunks[0]
assert top_match is not None
assert top_match.metadata["document_id"] == expected_doc_id, f"Query '{query}' should match {expected_doc_id}"
def test_insert_chunks_with_precomputed_embeddings(client_with_empty_registry, embedding_model_id):
vector_db_id = "test_precomputed_embeddings_db"
client_with_empty_registry.vector_dbs.register(
vector_db_id=vector_db_id,
embedding_model=embedding_model_id,
embedding_dimension=384,
)
chunks_with_embeddings = [
Chunk(
content="This is a test chunk with precomputed embedding.",
metadata={"document_id": "doc1", "source": "precomputed"},
embedding=[0.1] * 384,
),
]
client_with_empty_registry.vector_io.insert(
vector_db_id=vector_db_id,
chunks=chunks_with_embeddings,
)
# Query for the first document
response = client_with_empty_registry.vector_io.query(
vector_db_id=vector_db_id,
query="precomputed embedding test",
)
# Verify the top result is the expected document
assert response is not None
assert len(response.chunks) > 0
assert response.chunks[0].metadata["document_id"] == "doc1"
assert response.chunks[0].metadata["source"] == "precomputed"