NeuML

/

colbert-bert-tiny

Like 2

Follow

NeuML 175

Hello!

As of Sentence Transformers v6.0.0, this checkpoint loads directly as a multi-vector (ColBERT-style late interaction) retriever through the new MultiVectorEncoder. This PR adds a usage section to the model card and the multi-vector and sentence-transformers tags. The weights and the existing usage are untouched. Given the model's stated purpose, it might be handy that this also makes it a natural unit-test checkpoint for MultiVectorEncoder pipelines.

from sentence_transformers import MultiVectorEncoder

model = MultiVectorEncoder("NeuML/colbert-bert-tiny")

query = "What is the capital of France?"
documents = [
    "Paris is the capital and largest city of France.",
    "Berlin is the capital of Germany.",
]

query_embeddings = model.encode_query(query)
document_embeddings = model.encode_document(documents)
print(query_embeddings.shape, document_embeddings[0].shape)
# torch.Size([32, 128]) torch.Size([12, 128])

# MaxSim late-interaction scoring (higher is more relevant)
scores = model.similarity(query_embeddings, document_embeddings)
print(scores)
# tensor([[25.9327, 23.9168]], device='cuda:0')

Verified against a PyLate reference: the snippet reproduces exactly, and the token embeddings match with per-token cosine similarity above 0.999 and matching MaxSim scores. For reference, loaded through this integration the model scores 0.4035 mean nDCG@10 on NanoBEIR.

• Tom Aarsen