Commit History

  1. Enable speculattive decoding

    a7fdfe6
  2. fix role disorder error in history

    06a162a
  3. Add internet search feature

    4e60755
  4. fix reasonning model's thought process display

    9d3ca6c
  5. bugfix

    4522453
  6. fix for multi-turn conv.

    6c77ec7
  7. bugfix for think tag handling

    14564aa
  8. support reasoning tag

    5db22d5
  9. add missing part of app.py

    afa19a3
  10. add 4 new models

    d554072
  11. fix error : ValueError: Conversation roles must alternate user/assistant/user/assistant

    3e4847c
  12. fault-free model loading

    6e8312c
  13. add storage indicator

    cc91a1a
  14. improve model management

    0813164
  15. improve storage management

    37ee1f3
  16. provide more models, secure memory usage

    cd26609
  17. increase context length to 2048

    56919fd
  18. switch to 7b model

    ea7adad

  1. switch to 3b q2 model

    88b1f39
  2. adjust thread numbers

    4443d46
  3. switch to 1.5b

    4d633ef
  4. fix model id

    8287454
  5. Add app.py & requirements.txt

    0ff6c39