guoch/bce-embedding-base_v1-openvino-qint8

Model

OpenVINO Version 8bit of bce-embedding-base_v1. Using: https://huggingface.co/spaces/sentence-transformers/backend-export. Calibrated using params: nyu-mll/glue, sst2, train, sentence, 300

Inference

from sentence_transformers import SentenceTransformer

# 1. Load the model from the HF中国镜像站 Hub
model = SentenceTransformer(
    "guoch/bce-embedding-base_v1-openvino-qint8",
    backend="openvino",
    model_kwargs={"file_name": "openvino_model_qint8_quantized.xml"},
)

# 2. Inference works as normal
embeddings = model.encode(["The weather is lovely today.", "It's so sunny outside!", "He drove to the stadium."])
similarities = model.similarity(embeddings, embeddings)