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 10, 14:00 UTC
main
[Refactor][TPU] Remove torch_xla path and use tpu-inference (#30808)
weiyu committed 10 months ago
[Bugfix][Kernel] fix bias adding in triton kernel implemented fused moe (#31676)
xuebwang-amd committed 10 months ago
[Bugfix][Hardware][AMD] Consolidate FP8 min/max values helper function (#31106)
Kevin McKay committed 10 months ago
[BugFix] LoRA: Support loading base_layer of experts (#31104)
ℍ𝕠𝕝𝕝𝕠𝕨 𝕄𝕒𝕟 committed 10 months ago
[Bugfix] Fix race condition in async-scheduling for vlm model (#31841)
tianshu-Michael-yu committed 10 months ago
refactor: find_loaded_library (#31866)
tjp_zju committed 10 months ago
Lucas Wilkinson committed 10 months ago
[ROCm][AITER] bugfix accuracy regression in ROCM_AITER_TRITON_MLA backend (#31816)
vllmellm committed 10 months ago
[Chore] Try remove `init_cached_hf_modules` (#31786)
Cyrus Leung committed 10 months ago
fixed mypy warnings for files vllm/v1/attention with TEMPORARY workaround (#31465)
Jack Yang committed 10 months ago
Change warning in get_current_vllm_config to report caller's line number (#31855)
Tyler Michael Smith committed 10 months ago
[Doc] Update release docs (#31799)
Cyrus Leung committed 10 months ago
[Model] Enable LoRA support for PaliGemma (#31656)
Ce Zhao committed 10 months ago
[1/2][lmcache connector] clean up lmcache multi-process adapter (#31838)
Yihua Cheng committed 10 months ago
[Misc][BE] Type coverage for vllm/compilation [1/3] (#31554)
Lucas Kabela committed 10 months ago
[Frontend] Implement robust video frame recovery for corrupted videos (#29197)
vSeamar committed 10 months ago
Andreas Karatzas committed 10 months ago
[CI] Add warmup run in test_fusion_attn (#31183)
Angela Yi committed 10 months ago
[Bugfix] Handle mistral tokenizer in get_hf_processor (#31817)
Cyrus Leung committed 10 months ago
[ROCm][CI] Pinning timm lib version to fix ImportError in Multi-Modal Tests (Nemotron) (#31835)
Andreas Karatzas committed 10 months ago