Commit History

Autor SHA1 Mensaxe Data
  AlpinDale 696f2cd59c add phi3_small support with blocksparse attention hai 6 meses
  AlpinDale 1b86cf6164 navi21 fallback to naive attention hai 6 meses
  AlpinDale 0c15965621 fix fp8 kv hai 6 meses
  AlpinDale a94de94c44 refactor: combine the prefill and decode into a single API (#553) hai 6 meses
  AlpinDale 50b7c13db0 refactor: attention selector (#552) hai 6 meses
  AlpinDale 8b56dc4347 dict -> torch.Tensor for blocks_to_swap hai 6 meses
  AlpinDale 3a0d1c7705 add get_name method to attention backends hai 6 meses
  AlpinDale 21ce19b3ea blocks_to_copy dict -> torch.Tensor hai 6 meses
  AlpinDale 35ae01d7ba refactor: attention metadata term hai 6 meses
  AlpinDale aa15d3dc0f sliding window in prefix prefill kernel hai 6 meses
  AlpinDale 0d3562a7f9 MQA in triton FA hai 6 meses
  AlpinDale 0f7ef9ef7c fix: import in selector hai 6 meses
  AlpinDale 46159b107a formatting: pt1 hai 7 meses
  AlpinDale fca911ee0a vLLM Upstream Sync (#526) hai 7 meses
  AlpinDale 66b7bc4415 sliding window in prefix kernel hai 8 meses
  AlpinDale 9d81716bfd [v0.5.3] Release Candidate (#388) hai 8 meses