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 5, 13:30 UTC
main
[vLLM IR] add `import_ir_kernels()` to support OOT platforms (#38807)
Xiaoshuang Wang committed 7 months ago
Fix invalid logprobs with MTP enabled and sync scheduling (#38711)
danisereb committed 7 months ago
[Frontend] new online quantization frontend (#38138)
Vasiliy Kuznetsov committed 7 months ago
[KVConnector] Skip `register_kv_caches` on profiling (#38558)
Nicolò Lucchesi committed 7 months ago
[Bugfix] Fix AWQ models batch invariance issues (#38670)
Yusuf Mohammad committed 7 months ago
[XPU] bump up xpu-kernel v0.1.5, transpose moe weights (#38342)
Qiming Zhang committed 7 months ago
[Kernel] Add swapAB support for SM120 CUTLASS blockwise FP8 GEMM (#38325)
Necofish committed 7 months ago
[GDN] Eliminate GPU->CPU sync in prepare_chunk_indices during prefill (#38361)
Artem Perevedentsev committed 7 months ago
[Intel][Triton] Support `round_int8` for Intel backend (#38825)
Mieszko Dziadowiec committed 7 months ago
[XPU][CI] Skip test_topp_only and test_topk_and_topp cases on Intel GPU in CI (#38904)
xiangdong committed 7 months ago
[ROCm] Fix aiter persistent mode mla with q/o nhead<16 for kimi-k2.5 tp8 (#38615)
wufann committed 7 months ago
[CPU] Replace OMP initialization (#36487)
Anton Ivanov committed 7 months ago
[Frontend] feat: add streaming support for token generation endpoint (#37171)
Hyeonki Hong committed 7 months ago
[XPU][CI] Skip test_topk_only cases on Intel GPU in CI (#38899)
xiangdong committed 7 months ago
[ROCm][CI/Build] Fix the pytest hook to properly print out the summary (#38585)
Gregory Shtrasberg committed 7 months ago
Fix Nano Nemotron VL regressions (#38655)
Netanel Haber committed 7 months ago
[Misc] Clean up Gemma4 implementation (#38872)
Isotr0py committed 7 months ago
[MRV2][KVConnector] Fix missing build_connector_worker_meta (#38698)
Yifan Qiao committed 7 months ago
[CI/Build] Add audio deps in Dockerfile.cpu (#38876)
Li, Jiang committed 7 months ago
[Bug] Add e_score_correction_bias to SKIP_TENSORS (#38746)
Aaron Hao committed 7 months ago