Commit Verlauf

Autor SHA1 Nachricht Datum
  AlpinDale 45a004874c chore: allow specifying custom Executor vor 5 Monaten
  AlpinDale a26f784240 chore: use the LoRA tokenizer in OpenAI API (#599) vor 5 Monaten
  AlpinDale 052a6e1eb6 feat: add SPMD worker execution using Ray accelerated DAG vor 5 Monaten
  AlpinDale 0c17c2a8a7 chore: add commit hash, clean up engine logs vor 5 Monaten
  AlpinDale c0c2b1ac20 fix: get_and_reset only when scheduler outputs are not empty vor 5 Monaten
  AlpinDale 99680b2d23 feat: soft prompts (#589) vor 5 Monaten
  AlpinDale 4f7d212b70 feat: remove vision language config vor 5 Monaten
  AlpinDale 5be90c3859 Mamba infrastrucuture support (#586) vor 5 Monaten
  AlpinDale ae04f57ec1 feat: Pipeline Parallel support (#581) vor 5 Monaten
  AlpinDale 9da2448964 fix: ensure worker model loop is always stopped at the right time vor 5 Monaten
  AlpinDale 7b04361934 fix: support getting `eos_token_id` from the config file vor 5 Monaten
  AlpinDale b8a19ba27f chore: extend aphrodite metrics logging api vor 5 Monaten
  AlpinDale 0886c361f4 feat: OpenVINO CPU backend (#576) vor 5 Monaten
  AlpinDale c0c336aaa3 refactor: registry for processing model inputs; quick_gelu; clip model support vor 5 Monaten
  AlpinDale b3643a7bd7 fix: min_tokens for when there are multiple eos tokens vor 5 Monaten
  AlpinDale 4ed1bb9958 chore: add fault tolerance for RayTokenizerGroupPool vor 5 Monaten
  AlpinDale 25feb1d592 chore: add support for pinning lora adapters in the lru cache vor 5 Monaten
  AlpinDale 6a57861fca feat: initial XPU support via intel_extension_for_pytorch (#571) vor 6 Monaten
  AlpinDale a07fc83bc8 chore: proper util for aphrodite version vor 6 Monaten
  AlpinDale fe21123a1c feat: TPU support (#570) vor 6 Monaten
  AlpinDale 90ceab32ff refactor: consolidate prompt args to LLM engines vor 6 Monaten
  AlpinDale de62ceb18c refactor: eliminate parallel worker per-step task scheduling overhead vor 6 Monaten
  AlpinDale 60e74e92fd add rope_scaling arg vor 6 Monaten
  AlpinDale c6a501f682 add multiprocessing executor; make ray optional vor 6 Monaten
  AlpinDale 342346afda improve hashing function vor 6 Monaten
  AlpinDale 50b7c13db0 refactor: attention selector (#552) vor 6 Monaten
  AlpinDale fd0a5c0ea4 raise a warning during preemption and swapping vor 6 Monaten
  AlpinDale be8154a8a0 feat: proper embeddings API with e5-mistral-7b support vor 6 Monaten
  AlpinDale ef733aee43 implement ExecuteModelData to reduce executor complexity vor 6 Monaten
  AlpinDale ba3db54a4b comment out the chunked debug print vor 6 Monaten