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 12, 23:30 UTC
main
[Bugfix] Fix FusedMoEModularKernel for triton backend (#28913)
Xin Yang committed 12 months ago
[Attention] FlashAttention ViT support, make default backend (#28763)
Matthew Bonanni committed 12 months ago
Enable bitsandbytes quantization on AMD GPUs that use warp size 32 (#27307)
Strahinja Stamenkovic committed 12 months ago
[CPU] Refactor CPU WNA16 (#28826)
Li, Jiang committed 12 months ago
[Core] Switch Flat logprob control from environment variable to SamplingParams (#28914)
Jialin Ouyang committed 12 months ago
Move online quantization to `model.load_weights` (#26327)
Jerry Zhang committed 12 months ago
tomeras91 committed 12 months ago
[Build] Add OpenAI triton_kernels (#28788)
Varun Sundar Rabindranath committed 12 months ago
[NVIDIA] Guard SM100 CUTLASS MoE macro to SM100 builds v2 (#28938)
Johnny committed 12 months ago
Supress verbose logs from model_hosting_container_standards (#28949)
Michael Goin committed 12 months ago
Replace `torch.cuda.Event` with `torch.Event` for better hardware compatibility (#26985)
Kunshang Ji committed 12 months ago
[NIXL] fix cpu PD after physical <> logical block_size PR (#28904)
Chendi.Xue committed 12 months ago
[Models] Replace all `nn.Conv2d` with vLLM's Conv2dLayer (#28842)
Isotr0py committed 12 months ago
[chore] Move the rest of wikimedia url to S3 (#28921)
Kevin H. Luu committed 12 months ago
[FEAT] [AITER] [ROCm] integrate aiter sampling ops (#26084)
vllmellm committed 12 months ago
[Minor] Rename `ec_producer` field to `is_ec_producer` (#28884)
Nick Hill committed 12 months ago
[CI][NIXL] Change default `block_size` for tests (#28927)
Nicolò Lucchesi committed 12 months ago
[Model] Add Gemma3 GGUF multimodal support (#27772)
Luciano Martins committed 12 months ago
[Benchmark] multi_turn: Report warmup-inclusive runtime (#28937)
Ido Segev committed 12 months ago
[CI Sprint] Quantization CI Cleanup (#24130)
Alex committed 12 months ago