Historique des commits

Auteur SHA1 Message Date
  AlpinDale e0886ee929 feat: add `ProposerWorkerBase` abstract class il y a 7 mois
  AlpinDale 344ddaac5a properly disable speculative decoding il y a 8 mois
  AlpinDale 5b0c11d190 support pipeline parallel pynccl groups il y a 8 mois
  AlpinDale de62ceb18c refactor: eliminate parallel worker per-step task scheduling overhead il y a 8 mois
  AlpinDale 236be273e5 feat: tensor parallel speculative decoding (#554) il y a 8 mois
  AlpinDale 197a6d2c16 auto disable speculative decoding by the running queue size il y a 8 mois
  AlpinDale 438f5bdce9 fix ngrams il y a 8 mois
  AlpinDale ef733aee43 implement ExecuteModelData to reduce executor complexity il y a 8 mois
  AlpinDale 79901b76de logprobs for target model (spec decoding) il y a 8 mois
  AlpinDale 723c6acb84 re-add ngram speculative decoding il y a 8 mois
  AlpinDale 7bcf4c3fc9 centralize gpu worker construction il y a 8 mois
  AlpinDale 9d81716bfd [v0.5.3] Release Candidate (#388) il y a 10 mois
  AlpinDale f8dfac6372 chore: attention refactor and upstream sync apr01 (#365) il y a 1 an