A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 5, 18:30 UTC
main
[WIP][CI][Bugfix] Fix `test_run_eagle_dp` (#38584)
Matthew Bonanni committed 7 months ago
Mateusz Sokół committed 7 months ago
[Misc] Move --grpc CLI argument into make_arg_parser (#38570)
Chang Su committed 7 months ago
[Feature]: add presence_penalty and frequency_penalty fields to Responses API (#38613)
Chauncey committed 7 months ago
[Frontend][3/n] Improve pooling entrypoints | scoring. (#28631)
wang.yuqi committed 7 months ago
[CPU] Support int8 compute mode in CPU AWQ (#35697)
Yintong Lu committed 7 months ago
[ci] Remove benchmarks job (#38611)
Kevin H. Luu committed 7 months ago
[OOT] Add OOT support for linear kernel. (#37989)
zhangyiming committed 7 months ago
Kfir Toledo committed 7 months ago
[Tool Parser][2/3] Use self.tools instead of request.tools in tool parsers (#38189)
Flora Feng committed 7 months ago
vLLM Benchmark Suite perf regression after PR#32723 (#38576)
Louie Tsai committed 7 months ago
[ROCm][CI] Fix Whisper translation test attention backend selection (#38508)
Andreas Karatzas committed 7 months ago
[Mypy] Fix adjust_request typing (#38264)
Flora Feng committed 7 months ago
[KVConnector] Remove redundant method KVConnectorOutput::merge() (#38546)
Martin Hickey committed 7 months ago
[Bugfix] Fix for builtins (forward fix of pytorch/177558) (#37234)
Lucas Kabela committed 7 months ago
[Refactor] Consolidate Tool type alias in tool_parsers/utils.py (#38265)
Flora Feng committed 7 months ago
[Bugfix][DCP] Fix CUDA graph capture for Decode Context Parallelism (#36070)
sungsoo ha committed 7 months ago
Add @vadiklyutiy as committer (#38589)
Vadim Gimpelson committed 7 months ago
feat(attention): extract KV-cache update from FlashAttentionDiffKV ba… (#36466)
Prathmesh Bhatt committed 7 months ago
Netanel Haber committed 7 months ago