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, 02:15 UTC
main
[Bug] Fix batch invariant test issue, bs=1 with `max_seq_num = 1` (#39320)
Wentao Ye committed 7 months ago
[Speculative Decoding] Add DFlash speculators config parsing (#38300)
zhanqiuhu committed 7 months ago
[Bugfix][NIXL] Fix `_logical_to_kernel_block_ids` conversion for non-mamba models (#39724)
zhanqiuhu committed 7 months ago
[CI] Only build release Docker images when NIGHTLY=1 (#39882)
Kevin H. Luu committed 7 months ago
[Mooncake] Fix mixed MLA+Eagle block-size validation (#39596)
Zhewen Li committed 7 months ago
Monishver committed 7 months ago
[Test] Fix @create_new_process_for_each_test("fork") in interactive shell pipeline (#29130)
Mark McLoughlin committed 7 months ago
[BugFix] KeyError on scope["method"] for realtime api websocket in AuthenticationMiddleware (#36934)
daniebrill committed 7 months ago
Roy Huang committed 7 months ago
[SpecDecode][Benchmark] Add SPEED-bench support to benchmarking CLI (#36029)
Talor Abramovich committed 7 months ago
[XPU] properly handle q_descale on XPU as quant query input not supported (#39676)
Yan Ma committed 7 months ago
FIX: support language_model.backbone naming in NemotronH Nano VL quantization config (#39901)
danielafrimi committed 7 months ago
[BugFix][Graph] fix: handle empty sym_shape_indices in PiecewiseBackend. (#39395)
Chauncey committed 7 months ago
[XPU][MXFP4] add mxfp4 quant op for XPU (#39857)
zofia committed 7 months ago
Csrayz committed 7 months ago
[Quantization][Autoround][CPU] Add W4A16 Support (#38192)
Zhenzhong Xu committed 7 months ago
fix online fp8 for MiniCPM models (#39862)
Yan Ma committed 7 months ago
[kv_offload+HMA][3/N]: Remove block_size from KVEvents (#36644)
Or Ozeri committed 7 months ago
[Bugfix] Fix tool_calls Iterable consumed when debug logging is enabled (#34844)
Wojciech Wais committed 7 months ago
[CI][NIXL] Fix PD CI breakage: pin nixl-cu{12,13} versions (#39851)
zhanqiuhu committed 7 months ago