Histórico de Commits

Autor SHA1 Mensagem Data
  AlpinDale 00503b9fc1 feat: non-uniform quantization via `compressed-tensors` for llama há 5 meses atrás
  AlpinDale ee2c5d34da feat: add fp8 channel-wise weight quantization support há 5 meses atrás
  AlpinDale 98cb1c4cd1 feat: support fp8 via `llm-compressor` há 5 meses atrás
  AlpinDale e2dbe5f05c feat: add sparse marlin for compressed tensors há 6 meses atrás
  AlpinDale aba03b4756 feat: dynamic per-token activation quantization há 6 meses atrás