História revízii

Autor SHA1 Správa Dátum
  AlpinDale 058e629f8e chore: refactor marlin python utils 6 mesiacov pred
  AlpinDale c0c2b1ac20 fix: get_and_reset only when scheduler outputs are not empty 6 mesiacov pred
  AlpinDale b9268be8e8 fix: engine timeout due to request abort 6 mesiacov pred
  AlpinDale 8a44866e00 restrict outlines to < 0.1 6 mesiacov pred
  AlpinDale 4501ae5f15 fix: neuron executor for adapters 6 mesiacov pred
  AlpinDale 16dff9babc chore: enable bonus token in spec decoding for KV cache based models 6 mesiacov pred
  AlpinDale 4150b1ea3a fix: adapter methods for OpenVINO executor 6 mesiacov pred
  AlpinDale db73f03cdc fix: use ParallelLMHead for MLPSpeculator 6 mesiacov pred
  AlpinDale 9622c59f8f chore: support 2D input shape in MoE layer 6 mesiacov pred
  AlpinDale 4628caeae6 fix: missed these adapter methods for TPU executor 6 mesiacov pred
  AlpinDale dba22e4f83 fix: add zeromq fallback for broadcasting large objects (e.g. vlm images) 6 mesiacov pred
  AlpinDale d9f4c36edd feat: Medusa speculative decoding support (#590) 6 mesiacov pred
  AlpinDale 6abf4e3883 fix: needs_scalar_to_array logic check in linear layer 6 mesiacov pred
  AlpinDale a3b56353fa fix: another one missed 6 mesiacov pred
  AlpinDale 4e4cd55d30 fix: incorrect LoRA import 6 mesiacov pred
  AlpinDale 99680b2d23 feat: soft prompts (#589) 6 mesiacov pred
  AlpinDale 1cb06835a0 fix: TPU multimodal kwargs and outlines installation in TPU docker 6 mesiacov pred
  AlpinDale 1562e073c6 fix: ray worker rank assigment 6 mesiacov pred
  AlpinDale 1a40bf438b fix: incorrect gpu capability when used mixed gpus 6 mesiacov pred
  AlpinDale 3798ecc309 chore: add flashinfer to default dockerfile 6 mesiacov pred
  AlpinDale ebba0d9226 fix: mamba cache cuda graph padding 6 mesiacov pred
  AlpinDale c25a9abb28 fix: outlines failing on second launch 6 mesiacov pred
  AlpinDale 2105e4fd6b feat: correctly invoke prefill & decode kernels for cross-attention 6 mesiacov pred
  AlpinDale 3e7d5f7d14 chore: reloading fused_moe config on the last chunk 6 mesiacov pred
  AlpinDale 88a638d793 chore: debug logs for all available endpoints 6 mesiacov pred
  AlpinDale 98cb1c4cd1 feat: support fp8 via `llm-compressor` 6 mesiacov pred
  AlpinDale bf4f113ef1 feat: add paligemma vision model support 6 mesiacov pred
  AlpinDale 7e99578712 fix: cleanup validation and update docs for vlm 6 mesiacov pred
  AlpinDale 526163003d fix: improve consistency between feature size calc and dummy data for profiling 6 mesiacov pred
  AlpinDale c11a8bdaad fix: calculate max number of multi-modal tokens automatically 6 mesiacov pred