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 9, 00:00 UTC
main
[Bugfix] Disable RoutingMethodType.[Renormalize,RenormalizeNaive] TRTLLM per-tensor FP8 MoE (#33620)
Michael Goin committed 9 months ago
[Refactor] Clean up pooling serial utils (#33665)
Cyrus Leung committed 9 months ago
[Bugfix][Model] Fix DeepSeek-OCR-2 chat template to include BOS token (#33642)
Lucas Hänke de Cansino committed 9 months ago
[CI/Build] Investigate torchrun distributed tests hanging issue (#33650)
Isotr0py committed 9 months ago
[torch.compile] Document the workaround to standalone_compile failing (#33571)
Richard Zou committed 9 months ago
[Misc] Remove deprecated VLLM_ALL2ALL_BACKEND environment variable (#33535)
杨朱 · Kiki committed 9 months ago
[Minor] Some code simplification in `scheduler.py` (#33597)
Nick Hill committed 9 months ago
[Misc] Remove deprecated profiler environment variables (#33536)
杨朱 · Kiki committed 9 months ago
[XPU][1/N] Deprecate ipex and switch to vllm-xpu-kernels for xpu platform (#33379)
Kunshang Ji committed 9 months ago
[Bugfix] Interleaved thinking keeps compatibility with reasoning_content (#33635)
Chauncey committed 9 months ago
[CI/Build] Remove hardcoded America/Los_Angeles timezone from Dockerfiles (#33553)
杨朱 · Kiki committed 9 months ago
Fix quantized Falcon-H1 model loading issues (#32728)
Shengliang Xu committed 9 months ago
[Frontend] Add sampling parameters to Responses API (#32609)
Daniel Mescheder committed 9 months ago
[Bugfix] Fix mm budget setting for Qwen Omni models (#33634)
Roger Wang committed 9 months ago
[Feature][CPU Backend]: Optimize ARM vectorization backend (#30329)
Radu Salavat committed 9 months ago
Richard Zou committed 9 months ago
[CI/Build] add directions for CPU image upload to Docker Hub (#32032)
Nathan Weinberg committed 9 months ago
[BugFix] DPMetadata raises assert error for dense model (#32739)
Dezhan committed 9 months ago
[Voxtral Realtime] Introduce global log mel max (#33574)
Patrick von Platen committed 9 months ago
fix cutlass_3x_gemm_fp8_blockwise on sm103a (#32224)
Lain committed 9 months ago