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 13, 03:15 UTC
main
Cyrus Leung committed 12 months ago
[Performance][DeepGEMM] Estimate expected_m (#28694)
Varun Sundar Rabindranath committed 12 months ago
[NIXL] heterogeneous block_size support (#26759)
Chendi.Xue committed 12 months ago
Fix IntermediateTensors initialization and add type hints (#28743)
Mohammad Othman committed 12 months ago
Revert "[Core] Performance: Use list[np.ndarray] instead of list[list… (#28773)
Nick Hill committed 12 months ago
[Bugfix] Build hadacore kernels on >SM90 (#28748)
Michael Goin committed 12 months ago
[Model][Qwen3VL] Use `mm_position` to compute mrope positions (#28730)
Lukas Geiger committed 12 months ago
[TPU] Fix import error in tpu launch (#28758)
QiliangCui committed 12 months ago
Jialin Ouyang committed 12 months ago
[Test] Rework e2e async scheduling tests (#28744)
Nick Hill committed 12 months ago
[ROCm][CI/Build] Upgrade to ROCm 7.1 and AITER main (#28753)
Gregory Shtrasberg committed 12 months ago
[Log] Save profiler results to file instead of stdout (#28144)
rasmith committed 12 months ago
[Hybrid] [Kernel] Fix chunk scan kernel when BLOCK_SIZE_DSTATE > 128 (#28295)
Thomas Parnell committed 12 months ago
[Bugfix] Fix ChunkedLocalAttention CUDA Graph setting (#28739)
Benjamin Chislett committed 12 months ago
Alexander Matveev committed 12 months ago
Avoid bytecode hook and simplify TorchCompileWrapperWithCustomDipatch (#25110)
Laith Sakka committed 12 months ago
[ROCm][CI/Build] Change install location of uv (#28741)
Gregory Shtrasberg committed 12 months ago
[Bugfix] TypeError: 'NoneType' object is not callable (#27410)
Marcin Ostrowski committed 12 months ago
[CI] Fix macos smoke test uv cache issue (#28736)
Michael Goin committed 12 months ago
[ROCm][Bugfix] Fix compilation errors with fused_qknorm_rope_kernel.cu (#28682)
Sage Moore committed 12 months ago