SLM-RAG-Arena / utils /models.py

Commit History

  1. remove debugging

    93fe5e0
  2. format prompts and add debug log

    798ebc4

  1. remove cogito for gpu-hour reasons and add approved llama models

    e7fd2d9
  2. add additional models

    dc2d37a
  3. fix typo

    c0f44db
  4. gemma 3 4b misaligned bug, so removing that as an option for now

    a837a1c
  5. update model list

    533ae49
  6. revert to sequential processing

    97629be
  7. typo fix

    e6127a4
  8. prepare for zeroGPU

    b8ee0a2
  9. specified device

    5021e53
  10. vllm backend swap v1

    6c63a2d

  1. proper threaded generation interrupt

    6b26b26
  2. Handled interruption (#10)⁠

    oliver-aizip +1: kai-aizip
    Verified signature
    1db9e92
  3. ready hyperlinks for leaderboard

    44c2a20
  4. first pass at async handling

    665e5a3
  5. switched models to using full context

    8151596

  1. fix models

    dab8aab
  2. remove gated models for now

    cb092d7
  3. v1 inference code

    ddaff53
  4. add UI and data processing

    8a142a6