Skip to main content
Superlinked is a self-hosted inference engine (SIE) for embedding, reranking, and extraction. The sie-chroma package exposes SIE as a Chroma EmbeddingFunction, giving you access to 85+ dense and sparse text embedding models from a single endpoint. You need a running SIE instance; see the Superlinked quickstart for deployment options.
Install the sie-chroma package:
pip install sie-chroma
Use SIEEmbeddingFunction for dense embeddings:
import chromadb
from sie_chroma import SIEEmbeddingFunction

embedding_function = SIEEmbeddingFunction(
    base_url="http://localhost:8080",
    model="BAAI/bge-m3",
)

client = chromadb.Client()
collection = client.create_collection(
    name="documents",
    embedding_function=embedding_function,
)

collection.add(
    documents=[
        "Machine learning is a subset of artificial intelligence.",
        "Neural networks are inspired by biological neurons.",
        "Deep learning uses multiple layers of neural networks.",
    ],
    ids=["doc1", "doc2", "doc3"],
)

results = collection.query(query_texts=["What is deep learning?"], n_results=2)
For hybrid search on Chroma Cloud, SIESparseEmbeddingFunction returns learned sparse vectors (SPLADE / BGE-M3) as dict[int, float]:
from sie_chroma import SIESparseEmbeddingFunction

sparse_ef = SIESparseEmbeddingFunction(
    base_url="http://localhost:8080",
    model="naver/splade-v3",
)

​
Multimodal

Chroma’s EmbeddingFunction protocol accepts text input only. For image embedding with SIE-supported multimodal models (CLIP, SigLIP, ColPali), use the SIE SDK directly to pre-compute embeddings and pass them to Chroma via collection.add(embeddings=...): 
from sie_sdk import SIEClient
from sie_sdk.types import Item
import chromadb

sie = SIEClient("http://localhost:8080")
chroma = chromadb.Client()
collection = chroma.create_collection("images")

results = sie.encode(
    "openai/clip-vit-large-patch14",
    [Item(images=["img1.jpg"]), Item(images=["img2.jpg"])],
    output_types=["dense"],
)

collection.add(
    ids=["img1", "img2"],
    embeddings=[r["dense"].tolist() for r in results],
    metadatas=[{"path": "img1.jpg"}, {"path": "img2.jpg"}],
)