sie-weaviate
SIE integration for Weaviate v4.
Two integration paths
1. Client-side (this package, works now)
sie-weaviate provides vectorizer and enrichment helpers that call SIE's
encode() and extract() and return data in the format Weaviate expects. You
configure collections with Configure.Vectors.self_provided() and pass
vectors on insert/query.
pip install sie-weaviate
import weaviate
import weaviate.classes as wvc
from sie_weaviate import SIEVectorizer
vectorizer = SIEVectorizer(base_url="http://localhost:8080", model="BAAI/bge-m3")
client = weaviate.connect_to_local()
try:
collection = client.collections.create(
"Documents",
properties=[wvc.config.Property(name="text", data_type=wvc.config.DataType.TEXT)],
vector_config=wvc.config.Configure.Vectors.self_provided(),
)
texts = ["first doc", "second doc"]
vectors = vectorizer.embed_documents(texts)
collection.data.insert_many([
wvc.data.DataObject(properties={"text": t}, vector=v)
for t, v in zip(texts, vectors)
])
query_vec = vectorizer.embed_query("search text")
results = collection.query.near_vector(near_vector=query_vec, limit=5)
finally:
client.close()
2. Server-side module (partnership, planned)
A text2vec-sie Go module for the Weaviate server that enables native
vectorizer config (Configure.Vectorizer.text2vec_sie(...)). See
weaviate-module-spec/ for the spec and reference implementation.
Named vectors (dense + multivector)
SIENamedVectorizer produces multiple vector types in one SIE call.
Use it with ColBERT models that output both dense and multivector
(per-token) embeddings:
from sie_weaviate import SIENamedVectorizer
vectorizer = SIENamedVectorizer(
base_url="http://localhost:8080",
model="jinaai/jina-colbert-v2",
output_types=["dense", "multivector"],
)
collection = client.collections.create(
"Documents",
properties=[wvc.config.Property(name="text", data_type=wvc.config.DataType.TEXT)],
vector_config=[
wvc.config.Configure.Vectors.self_provided(name="dense"),
wvc.config.Configure.Vectors.self_provided(name="multivector"),
],
)
named = vectorizer.embed_documents(["hello world"])
collection.data.insert_many([
wvc.data.DataObject(properties={"text": "hello world"}, vector=named[0])
])
For hybrid search, Weaviate has built-in BM25 — no extra vectors needed:
results = collection.query.hybrid(query="search text", alpha=0.75)
Document enrichment for Query Agent
SIEDocumentEnricher combines SIE's embedding and entity extraction
pipelines to produce documents with dense vectors and structured
metadata. The extracted properties (persons, organizations, locations,
categories) are exactly what Weaviate's Query Agent uses to construct
filters from natural language queries.
import weaviate
import weaviate.classes as wvc
from sie_weaviate import SIEDocumentEnricher
enricher = SIEDocumentEnricher(
base_url="http://localhost:8080",
labels=["person", "organization", "location"],
classify_model="knowledgator/gliclass-large-v3.0",
classify_labels=["technical", "business", "legal"],
)
client = weaviate.connect_to_local()
try:
collection = client.collections.create(
"Documents",
description="Documents with extracted entity and classification metadata.",
properties=[
wvc.config.Property(name="text", data_type=wvc.config.DataType.TEXT),
wvc.config.Property(
name="person", data_type=wvc.config.DataType.TEXT_ARRAY,
description="People mentioned in the document",
),
wvc.config.Property(
name="organization", data_type=wvc.config.DataType.TEXT_ARRAY,
description="Organizations mentioned in the document",
),
wvc.config.Property(
name="location", data_type=wvc.config.DataType.TEXT_ARRAY,
description="Locations mentioned in the document",
),
wvc.config.Property(
name="classification", data_type=wvc.config.DataType.TEXT,
description="Document category: technical, business, or legal",
),
wvc.config.Property(
name="classification_score", data_type=wvc.config.DataType.NUMBER,
description="Confidence score for the classification",
),
],
vector_config=wvc.config.Configure.Vectors.self_provided(),
)
# Embed + extract in one call
texts = [
"John Smith presented Google's new AI strategy in New York.",
"The court ruling on patent law affects tech companies.",
]
docs = enricher.enrich(texts)
collection.data.insert_many([
wvc.data.DataObject(properties=doc.properties, vector=doc.vector)
for doc in docs
])
# The Query Agent can now filter on extracted properties:
# "find documents about Google" → organization filter + vector search
# "show me legal documents mentioning John Smith" → classification + person filter
query_vec = enricher.enrich_query("AI strategy announcements")
results = collection.query.near_vector(near_vector=query_vec, limit=5)
finally:
client.close()
Testing
# Unit tests (no server needed)
pytest
# Integration tests (requires SIE + Weaviate)
pytest -m integration
1 maintainer
Metadata
1 maintainer
Release files for sie-weaviate 0.8.2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| sie_weaviate-0.8.2.tar.gz | 17.9 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| sie_weaviate-0.8.2-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 27.9 kB
Release files / sie_weaviate-0.8.2.tar.gz
| Download URL | sie_weaviate-0.8.2.tar.gz |
|---|---|
| Size | 17.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
397607179b739626cd417686fc992095a0e46558ca62482ebc3c4e3e8de3bd72
|
|
BLAKE2b-256 checksum How to use checksums |
68bfdd4f4cf2fb11d677e5953a4a73da168a66a81572f2f073aa721135773125
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 23, 2026.
Transparency logRelease files / sie_weaviate-0.8.2-py3-none-any.whl
| Download URL | sie_weaviate-0.8.2-py3-none-any.whl |
|---|---|
| Size | 10.0 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
99a5dcf0dd42db716596de5418d83a27d0af45e4035de0a2eca0754619d12b34
|
|
BLAKE2b-256 checksum How to use checksums |
c87bd1e21a8227c540fe09ac6f7ed2e611423a774e691ec63c2de8a1756d107a
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 23, 2026.
Transparency log
