Commit History

  1. model and quantized model both with attention

    Liman
    7c7c3b6

  1. add output_attentions true to config

    Liman
    fc60a15
  2. replace fp32 model with one with attention layers

    Liman
    1d87473
  3. onnx model with attention

    Liman
    25edbdb

  1. rename q8 model

    Liman
    2c3fa66
  2. add quantized models

    Liman
    95f2179

  1. config files

    Liman
    02fc9c9
  2. onnx model

    Liman
    66a59f9
  3. initial commit

    Verified signature
    4c2780c