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 14, 05:30 UTC
main
[V1][Metrics][Plugin] Add plugin support for custom `StatLoggerBase` implementations (#22456)
Tova Movshovitz committed about 1 year ago
[BugFix] Disable fp8 kv-cache by default for DeepSeek V3.2 (#27121)
Lucas Wilkinson committed about 1 year ago
[BugFix] fix graph partition signature (#27139)
Boyuan Feng committed about 1 year ago
[Chore] Separate out profiling utilities from vllm.utils (#27150)
dongbo910220 committed about 1 year ago
Lucas Wilkinson committed about 1 year ago
[Minor] Add some clarifying comments to recent changes (#27130)
Nick Hill committed about 1 year ago
Fix incorrect string formatting in barrier timeout exceptions (#27149)
Yongtao Huang committed about 1 year ago
[Chore] Clean up pytorch helper functions in `vllm.utils` (#26908)
Isotr0py committed about 1 year ago
[Models][QwenVL] Remove unnecessary `.contiguous()` calls (#27106)
Lukas Geiger committed about 1 year ago
[Misc] Refactor `get_kv_cache_spec` into `AttentionLayerBase` (#26587)
Nicolò Lucchesi committed about 1 year ago
[fix][cpu] fix prefill attention in CPU attention backend (#27035)
Fadi Arafeh committed about 1 year ago
[Feature] Batch Invariant: Support DeepGEMM and Blackwell (#27127)
Wentao Ye committed about 1 year ago
[Chore] Separate out `vllm.utils.mem_utils` (#27143)
iAmir97 committed about 1 year ago
[Test] Add test for /health endpoint on engine failure (#26074)
dongbo910220 committed about 1 year ago
[DOC][FEATURES][CPU]update cpu feature for v1 (#27135)
Chendi.Xue committed about 1 year ago
Varun Sundar Rabindranath committed about 1 year ago
[GPT-OSS] Structure_Tag support for gpt-oss tool-call in cot (#25515)
Hanchenli committed about 1 year ago
Huamin Li committed about 1 year ago
[Perf] Add H100 fused MoE config (#25398)
ZiTian Zhao committed about 1 year ago
[BugFix][Core] Fix error when enable async-scheduling in multi-node env (#25887)
Lehua Ding committed about 1 year ago