Commit History

Autor SHA1 Mensaxe Data
  AlpinDale db73f03cdc fix: use ParallelLMHead for MLPSpeculator hai 5 meses
  AlpinDale 0f4a9ee77b quantized lm_head (#582) hai 5 meses
  AlpinDale de7e6919c0 feat: support tied weights and input scale for MLPSpeculator hai 6 meses
  AlpinDale 51cfadeb29 fix: `MLPSpeculator` handling of `num_speculative_tokens` hai 6 meses
  AlpinDale af43576da0 feat: add MLPSpeculator speculative decoding support (#572) hai 6 meses