A demonstration model produced by the native model-merging feature (SentenceTransformer.merge) added to Sentence Transformers. It spherically interpolates (slerp) between the two models at t = 0.5.

Merged checkpoints:

How it was created

python

from sentence_transformers import SentenceTransformer

merged = SentenceTransformer.merge(
    models=["sentence-transformers/all-MiniLM-L6-v2",
            "sentence-transformers/multi-qa-MiniLM-L6-cos-v1"],
    weights=[0.5, 0.5],
    method="slerp",
    output_path="all-MiniLM-L6-v2-multi-qa-slerp-merge",
    dtype="float32",
)

Usage

python

from sentence_transformers import SentenceTransformer

model = SentenceTransformer("yjoonjang/all-MiniLM-L6-v2-multi-qa-slerp-merge")
emb = model.encode(["Model merging combines fine-tuned checkpoints.",
                    "It often beats each individual model."])
print(model.similarity(emb, emb))

License

Derivative of the two base models above; their licenses apply. See each base model's card.