Commit Verlauf

Autor SHA1 Nachricht Datum
  AlpinDale 2b85ffb1a5 chore: minor cleanups vor 4 Monaten
  AlpinDale 4d4e767838 ci: take one of fixing lint issues vor 4 Monaten
  AlpinDale 0e6c400b13 feat: re-add GGUF (#600) vor 4 Monaten
  AlpinDale edffcecc67 chore: add proper logging for spec decoding verification vor 4 Monaten
  AlpinDale d357341203 chore: add pipeline parallel support for Qwen vor 4 Monaten
  AlpinDale 98f9dbd734 feat: Triton Kernels for Punica (#613) vor 4 Monaten
  AlpinDale 07cc8a56bb fix: add nemotron to PP_SUPPORTED_MODELS vor 4 Monaten
  AlpinDale ea838abb6b fix: disable enforce_eager for bnb vor 4 Monaten
  AlpinDale fce2c2e304 fix: support ignore patterns in model loader vor 5 Monaten
  AlpinDale cb44c8daa8 feat: support FP8 KV Cache scales from compressed-tensors vor 5 Monaten
  AlpinDale ba371fbbbd feat: AWQ marlin kernels (#603) vor 5 Monaten
  AlpinDale a4cbcfe59f feat: disable logprob serialization to CPU for spec decode vor 5 Monaten
  AlpinDale 9be43994fe feat: fbgemm quantization support (#601) vor 5 Monaten
  AlpinDale 45a004874c chore: allow specifying custom Executor vor 5 Monaten
  AlpinDale b7a2d52e47 fix: allow using mp executor for pipeline parallel vor 5 Monaten
  AlpinDale 6671e3a162 feat: add CPU offloading support (#598) vor 5 Monaten
  AlpinDale ee2c5d34da feat: add fp8 channel-wise weight quantization support vor 5 Monaten
  AlpinDale 6c4c20652b feat: pipeline parallel support for mixtral vor 5 Monaten
  AlpinDale 5289c14b24 feat: Asymmetric Tensor Parallel (#594) vor 5 Monaten
  AlpinDale ddb28a80a3 fix: bump torch for rocm, unify CUDA_VISIBLE_DEVICES for cuda and rocm vor 5 Monaten
  AlpinDale 99680b2d23 feat: soft prompts (#589) vor 5 Monaten
  AlpinDale 5761ef8c35 feat: gemma-2 support vor 5 Monaten
  AlpinDale 1ff6d4c3d7 feat: support pipeline parallel on indivisible GPU count (#587) vor 5 Monaten
  AlpinDale 4f7d212b70 feat: remove vision language config vor 5 Monaten
  AlpinDale bdf1cc1aec fix: allow using custom all reduce when pp_size > 1 vor 5 Monaten
  AlpinDale 5240c0da23 fix: avoid unnecessary ray import warnings vor 5 Monaten
  AlpinDale 5be90c3859 Mamba infrastrucuture support (#586) vor 5 Monaten
  AlpinDale ae04f57ec1 feat: Pipeline Parallel support (#581) vor 5 Monaten
  AlpinDale dd378ea063 feat: MLPSpeculator with tensor parallel vor 5 Monaten
  AlpinDale 3a0fdf7b9b chore: remove `image_input_type` from VLM config vor 5 Monaten