Commit History

  1. Fix cache max_seq_len

    537a375

  1. Update to V1.4

    28c720a

  1. update to 1.2

    69e8a46

  1. Optimize graph

    Verified signature
    9bfe4ad

  1. fix prev token

    Verified signature
    e90b8b5
  2. optimize compile by removing if branch

    Verified signature
    75e9ff1

  1. Better init event waiting

    Verified signature
    1caffd8
  2. wait for init

    Verified signature
    a4dfb48
  3. Update model to large sft

    Verified signature
    12b4214

  1. Enable compile on A10G

    Verified signature
    662d788

  1. Init hf space integration

    Verified signature
    0a3525d