A high-throughput and memory-efficient inference and serving engine for LLMs
main
[Bugfix] Annotate MTP draft KV cache groups positionally on the hybrid grouping path (#55390)
Navjot Singh committed about 1 month ago
[Perf][Attention] Reduce GLM sparse MLA preparation overhead (#57458)
Summer Yang committed about 1 month ago
[Scheduler] Soften Long Prefill Tokens Threshhold (#57951)
Robert Shaw committed about 1 month ago
[CI] Split (H200) LM Eval Large Models into per-model jobs (#57965)
Thang Nguyen committed about 1 month ago
[Pooling] MRV2 pooling shutdown model ref (#57737)
Taneem Ibrahim committed about 1 month ago
[Frontend] Add reusable TP1 initialized-engine snapshots (#51360)
Nils Matteson committed about 1 month ago
[Bugfix][Kernel] Skip the fused silu-mul block-quant fast path when a swiglu clamp is set (#57984)
Garrett Goon committed about 1 month ago
[CI][Build] Harden triton-cpu sleef submodule fetch in CPU image build (#57871)
vllm-agent committed about 1 month ago
[Kimi-K3][AMD] Return KDA and MLA projection outputs directly (#50592)
yinfengLiu committed about 1 month ago
[Docker] Use zstd for CI images and offer a Docker Hub variant (#55608)
Nils Matteson committed about 1 month ago
[Bugfix][ROCm] Reject unsupported EP for monolithic AITER MXFP4 MoE (#57866)
Andreas Karatzas committed about 1 month ago
[Bugfix][MoE] Reject hash routing for unsupported monolithic backends (#57867)
Andreas Karatzas committed about 1 month ago
[Engram] Drop redundant VLLM_PLE_CPU_OFFLOAD env var (#57937)
Nicolò Lucchesi committed about 1 month ago
[ROCm][Perf] Route the fused shared-expert gate GEMM through the platform dispatcher (#54185)
Mikko Tukiainen committed about 1 month ago
[Docs] Add an Engram feature page explaining Engram usage in vLLM (#57910)
Nicolò Lucchesi committed about 1 month ago
[Feature] Add first-class KV hints request envelope for programmatic KV management (#53423)
Karen Chung committed about 1 month ago
[ROCm][CI] Use ROCm backend for DeepSeek V4.1 ViT test (#57931)
djramic committed about 1 month ago
[MM] Move get_dummy_processor_inputs into MM processor (#57967)
Cyrus Leung committed about 1 month ago
[Bugfix] unskip InternViT test for transformers v5 compatibility (#55767)
Sahil Patel committed about 1 month ago
[ROCm] Refactor tuned gemms (#55001)
afriedri committed about 1 month ago