This dataset provides OpenResearcher corpus embeddings generated from Qwen3-Embedding-8B for building an offline search engine.
This dataset contains pre-computed embedding indexes stored as pickle files. Each .pkl file contains a tuple of:
embeddings (numpy.ndarray): Dense vector representations… See the full description on the dataset page:
https://huggingface.co/datasets/OpenResearcher/OpenResearcher-Indexes.
