datasets.json

{
  "datasets": {
    "wiki-10m": {
      "description": "Wikipedia 768-dimensions, ~87.5 million vectors",
      "download-url": "https://data.rapids.ai/raft/datasets/wiki_all_10M/wiki_all_10M.tar",
      "preparation-commands": "tar -xvf wiki_all_10M.tar",
      "base_file": "base.10M.fbin",
      "base_checksum": "8a0154c198592dfe096b0a94cfc89424abb1b64720bf9e10b3dad353e09d1fb9",
      "query_file": "queries.fbin",
      "query_checksum": "57c88c301a6ba032855af3a54bdb847a972004e1ab671ce1bff790ee9c39c855",
      "ground_truth_file": "groundtruth.10M.neighbors.ibin",
      "ground_truth_checksum": "6bf31445010a4f1c3513aa758a221af560ff8e03fcdd5afaa319b72aac6685b8",
      "num_docs": 10000000,
      "vector_dimension": 768,
      "top_k_ground_truth": 100
    },
    "sift-1m": {
      "description": "1 million 128-dimensional SIFT vectors",
      "download-url": "ftp://ftp.irisa.fr/local/texmex/corpus/sift.tar.gz",
      "preparation-commands": "tar -xvf sift.tar.gz",
      "base_file": "sift/sift_base.fvecs",
      "base_checksum": "21f66e2975057b5728ba56de1c825bac4f4d89d596609ae985741c6242631816",
      "query_file": "sift/sift_query.fvecs",
      "query_checksum": "f7fc9be140accdfd64116c2fa2365ecdb69b8f084970c6b0532db5ff79ac8fdc",
      "ground_truth_file": "sift/sift_groundtruth.ivecs",
      "ground_truth_checksum": "2b71de0a8d5a83e6a84eec3e23fb8b611d8801dd9b3a6cd62f070ab65ea65f4f",
      "num_docs": 1000000,
      "vector_dimension": 128,
      "top_k_ground_truth": 1000
    }
  }
}