BGE Base EN v1.5 — ONNX

ONNX export of BAAI/bge-base-en-v1.5, a high-quality English embedding model. Maps sentences to 768-dimensional dense vectors using CLS pooling with L2 normalization.

Mirrored for use with inference4j, an inference-only AI library for Java.

Original Source

Repository: BAAI (ONNX by Xenova)
License: mit

Usage with inference4j

try (SentenceTransformerEmbedder model = SentenceTransformerEmbedder.builder()
        .modelId("inference4j/bge-base-en-v1.5")
        .poolingStrategy(PoolingStrategy.CLS)
        .normalize()
        .build()) {
    float[] embedding = model.encode("Hello, world!");
    System.out.println("Dimension: " + embedding.length); // 768
}

Model Details

Property	Value
Architecture	BERT Base (12 layers, 768 hidden)
Task	Sentence embeddings / semantic similarity
Output dimension	768
Pooling	CLS
Normalization	L2
MTEB average	63.55
Max sequence length	512
Original framework	PyTorch (HuggingFace Transformers)

License

This model is licensed under the MIT License. Original model by BAAI, ONNX export by Xenova.

Downloads last month: -; Downloads are not tracked for this model. How to track