@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only
Fork 0

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 6, 05:45 UTC

Code Commits 22,390 Issues Pulls Cloud Agent CICD
main
main

[MoE] Move PF Methods to Folder (#35927)

Robert Shaw committed 8 months ago

4383f1532e

[ROCm][CI] Make some duplicated tests optional so that they are only evaluated in our nightly (#37780)

Andreas Karatzas committed 8 months ago

6eedec6e36

[ROCm][CI] Added missing resampy dependency for MM audio tests (#37778)

Andreas Karatzas committed 8 months ago

ffc8531524

[ROCm][CI] get_cu_count was renamed to num_compute_units in #35042 (#37764)

Andreas Karatzas committed 8 months ago

6ecba840d7

[ROCm][CI] Fix MEGA_AOT_ARTIFACT fallback when PyTorch < 2.10.0 lacks AOT support (#37763)

Andreas Karatzas committed 8 months ago

3b06c55c78

[Perf] Optimize glm4.xv VIT (#37779)

Yang Liu committed 8 months ago

b050700462

[Bugfix] Handle libsndfile sf_error(NULL) race condition in audio fallback (#37782)

Andreas Karatzas committed 8 months ago

5dac719b2b

[CI] Skip ISAAC multimodal tests due to broken upstream HF model weights (#37781)

Andreas Karatzas committed 8 months ago

c862481c02

[ROCm][CI] Add large_gpu_mark to test_max_tokens_none for ROCm (#37717)

Andreas Karatzas committed 8 months ago

c86b17cfe6

[Bugfix] Fix pooling non-determinism from pinned prompt_lens aliasing (#37775)

Andreas Karatzas committed 8 months ago

66f927f205

[ROCm][CI] close missing quote in kernels/moe block in run-amd-test.sh (#37774)

Andreas Karatzas committed 8 months ago

e78bc74268

[MoE] Move FlashInfer CuteDSL experts into fused_moe/experts/ (#37759)

Robert Shaw committed 8 months ago

6b2fa3a762

[Quantization][Deprecation] Remove PTPC FP8 (#32700)

Robert Shaw committed 8 months ago

eeee5b262d

Revert "Consolidate AWQ quantization into single awq_marlin.py file" (#37768)

Robert Shaw committed 8 months ago

5ad0446572

Consolidate AWQ quantization into single awq_marlin.py file

Robert Shaw committed 8 months ago

8cc700dd6a

Add tensor IPC transfer mechanism for multimodal data (#32104)

Brandon Pelfrey committed 8 months ago

80b70884eb

[Perf] Add SM 10.3 (B300/GB300) all-reduce communicator tuning (#37756)

Mohammad Miadh Angkad committed 8 months ago

61e381dcf0

[Core] Enable allreduce fusion by default for SM 10.3 (B300/GB300) (#37755)

Mohammad Miadh Angkad committed 8 months ago

88f1b374f5

[Hybrid] calling get_mamba_groups() once at MambaCopyBuffers.create() (#37318)

Francesco Fusco committed 8 months ago

298e510848

[ROCm] Enable DeepEP ROCm as all2allbackend for AMD GPUs. (#34692)

Chaitanya Sri Krishna Lolla committed 8 months ago

3982bc2cd0
← Newer Page 364 of 1120 Older →