[executor] feat: add supports_pp attr to executors (#21786)

H committed over 1 year ago

[Misc] update doc comment for send (#22026)

Ning Xie committed over 1 year ago

Support CUTLASS NVFP4 (w4a4) for Blackwell Geforce GPUs (SM120) (#21309)

Roberto L. Castro committed over 1 year ago

[Benchmark] Support ready check timeout in `vllm bench serve` (#21696)

Ye (Charlotte) Qi committed over 1 year ago

[Misc] Add tensor schema test coverage for multimodal models (#21754)

Isotr0py committed over 1 year ago

[Fix] Fix llama4 modelopt weight loading error (#22107)

jiahanc committed over 1 year ago

[Misc] Bump ray to 2.48.0 (#22123)

Rui Qiao committed over 1 year ago

Revert "[compile][startup] Disable C++ compilation of symbolic shapes" (#22122)

Xiao committed over 1 year ago

[xpu]support moe models on XPU platform (#21643)

Yan Ma committed over 1 year ago

[V1] [Hybrid] Validate compatibility of attention backend batch reordering at init time (#21557)

Thomas Parnell committed over 1 year ago

[Frontend] Improve error message for too many mm items (#22114)

Cyrus Leung committed over 1 year ago

[Model] Mamba2 preallocate SSM output tensor to avoid d2d copy overhead (#21075)

Chih-Chieh Yang committed over 1 year ago

for glm-4.1V update (#22000)

Yuxuan Zhang committed over 1 year ago

[PERF] Use faster way of decode in tokenizer: avoid useless list-to-list conversion (#20000)

Vadim Gimpelson committed over 1 year ago

docs: remove deprecated disable-log-requests flag (#22113)

Roger Wang committed over 1 year ago

[Bugfix] Mamba2 remove bugged initial state condition in chunk scan (#22034)

Chih-Chieh Yang committed over 1 year ago

Fix test_kv_sharing_fast_prefill flakiness (#22038)

Yong Hoon Shin committed over 1 year ago

[Misc] Getting and passing ray runtime_env to workers (#22040)

Rui Qiao committed over 1 year ago

[FEAT][ROCm] Enable running Flash Attention as ViT attn backend for Qwen-VL models on ROCm platform. (#22069)

vllmellm committed over 1 year ago

[Attention][DBO] Add support for "splitting" the CommonAttentionMetadata (#21153)

Sage Moore committed over 1 year ago