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 13, 22:00 UTC
main
feat(benchmarks): support HF model names in multi-turn benchmark (#27850)
ai-jz committed about 1 year ago
[Add] cmdline argument parsing for KV cache offloading modules (#27621)
Yihua Cheng committed about 1 year ago
[Bugfix] [Model] Missing MRoPE function definition from `KeyeForConditionalGeneration` (#27895)
TJian committed about 1 year ago
[CI/Build] Bump transformers version (#27528)
Cyrus Leung committed about 1 year ago
Adding SplitK in fused_moe_lora kernel (#27818)
yugong333 committed about 1 year ago
[Multimodal][XPU]Enable vision attn backend for xpu platform (#27525)
Yan Ma committed about 1 year ago
[Bugfix] Fix KDA output (#27905)
Jee Jee Li committed about 1 year ago
[Kernel] Enable FusedMoEModularKernel support bias (#27754)
Jee Jee Li committed about 1 year ago
[Core] Async scheduling + structured outputs compatibility (#26866)
Nick Hill committed about 1 year ago
[Hybrid] A simpler algorithm to find kernel_block_size (#26476)
Chen Zhang committed about 1 year ago
Bram Wasti committed about 1 year ago
Add FLASHINFER_MLA to test_mla_backends and add B200 CI run (#27663)
Matthew Bonanni committed about 1 year ago
[Feature] Pydantic validation for scheduler.py and structured_outputs.py (#26519)
Vinay R Damodaran committed about 1 year ago
[Cleanup] Remove no-longer-used `SpeculativeConfig.enable_chunked_prefill` (#27826)
Nick Hill committed about 1 year ago
Flashinfer_CUTLASS_MOE fuses quantization for TP (#27223)
Shu Wang committed about 1 year ago
fix incorrect type annotation in KimiMLP (#27885)
ZiTian Zhao committed about 1 year ago
[bugfix] Missing cached item in beam search (#27874)
Chenguang Zheng committed about 1 year ago
Docs update tpu install instructions (#27824)
Rob Mulla committed about 1 year ago
[Bugfix] Missing NIXL metadata for handshake initialization if instance spans multi-node (#26338)
GuanLuo committed about 1 year ago
[Bugfix] Fix broken MRoPE for GLM-4.1V/GLM-4.5V (#27860)
Isotr0py committed about 1 year ago