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 2, 06:00 UTC
main
[XPU][Bugfix] Fix per_token_group_fp8_quant missing dummy args on XPU (#43930)
Chaojun Zhang committed 5 months ago
[BugFix][kv_offload]: Prevent offloading stale sliding window blocks (#42959)
Or Ozeri committed 5 months ago
[BugFix] Fix TypeError in MiniCPM-O audio feature unpadding (#38053)
Krishna Chaitanya committed 5 months ago
[CI] Stabilize OpenAI schema fuzzing for malformed structural tags (#44131)
Andreas Karatzas committed 5 months ago
[Refactor] Move unstreamed tool-arg flush from serving layer to parser (#44017)
Flora Feng committed 5 months ago
[Bugfix][CI] Normalize NIXL connector CUDA wheel installs (#44266)
Alec committed 5 months ago
[ROCm] Upgrade AITER to v0.1.13.post1 (#44265)
Micah Williamson committed 5 months ago
[Bugfix][Mooncake] Release GPU pin on failed store in MooncakeStoreConnector (#43742)
Dao007forever committed 5 months ago
[DSV4] Refactor RoPE initialization (#44262)
Woosuk Kwon committed 5 months ago
[ROCm][CI] Skip unbacked dynamic shapes tests on PyTorch < 2.11 (#44256)
JartX committed 5 months ago
[DSV4] Remove unncessary classes & functions (#44246)
Woosuk Kwon committed 5 months ago
[Test][BugFix] Fix double-BOS in PD+specdec acceptance test (#44234)
Nick Hill committed 5 months ago
[BugFix][CI] Fix added `_has_module` tests (#44248)
Nick Hill committed 5 months ago
[Frontend][Core] Add sparse NCCL weight transfer support for in-place updates (#40096)
Siddharth Bedekar committed 5 months ago
[Rust Frontend] Support streaming `generate` endpoint (#43779)
Xunzhuo committed 5 months ago
[ROCm][CI] Fix and stabilize EAGLE3 acceptance tests (#41294)
Andreas Karatzas committed 5 months ago
[Kernel][DSv4] Optimize sparse FP8 compressor kernels (#44161)
Yongye Zhu committed 5 months ago
[Feature] Add support for JetBrains' Mellum v2 code generation model (#43992)
Madeesh Kannan committed 5 months ago
[Perf] Optimize cutlass fp8 scaled mm bypassing padding, 20% kernel performance improvement (#43706)
Wentao Ye committed 5 months ago
[XPU][CI] Fix test_audio_in_video flake by using module-scoped server fixture (#44146)
Chaojun Zhang committed 5 months ago