chibanaryan

minilm-pubmed-onnx

chibanaryan/minilm-pubmed-onnx

minilm-pubmed-onnx is a machine learning model from chibanaryan. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.

all-MiniLM-L6-v2 exported to ONNX and dynamically quantized to INT8. 5.3x faster than PyTorch on CPU (0.84ms vs 4.4ms per query), NDCG@5 -0.017 on the PubMed eval set. Mean-pool + L2-normalize the outputs. Serving mod…