bge-base-en-v1.5-gguf

Source model: https://huggingface.co/BAAI/bge-base-en-v1.5

Quantized and unquantized embedding models in GGUF format for use with llama.cpp. A large benefit over transformers is almost guaranteed and the benefit over ONNX will vary based on the application, but this seems to provide a large speedup on CPU and a modest speedup on GPU for larger models. Due to the relatively small size of these models, quantization will not provide huge benefits, but it does generate up to a 30% speedup on CPU with minimal loss in accuracy.

Files Available

Filename	Quantization	Size
bge-base-en-v1.5-f32.gguf	F32	417 MB
bge-base-en-v1.5-f16.gguf	F16	209 MB
bge-base-en-v1.5-q8_0.gguf	Q8_0	113 MB
bge-base-en-v1.5-q4_k_m.gguf	Q4_K_M	66 MB

Usage

These model files can be used with pure llama.cpp or with the llama-cpp-python Python bindings

from llama_cpp import Llama
model = Llama(gguf_path, embedding=True)
embed = model.embed(texts)

Here texts can either be a string or a list of strings, and the return value is a list of embedding vectors. The inputs are grouped into batches automatically for efficient execution. There is also LangChain integration through langchain_community.embeddings.LlamaCppEmbeddings.