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 11, 10:00 UTC
main
[Bugfix] Fix ViT with FlashAttention on ROCm (#30703)
Matthew Bonanni committed 11 months ago
Isotr0py committed 11 months ago
Remove `SkipValidation` from `ModelConfig` (#30695)
Harry Mellor committed 11 months ago
[Bugfix] Fix missing first token in tool calls during reasoning-to-tool transition (#30671)
mondaylord committed 11 months ago
[Frontend] add tools for dsv32 developer role (#30040)
yjc9696 committed 11 months ago
[MoE][Refactor 1/N] Separate Online Quantization (#30627)
Robert Shaw committed 11 months ago
[Bugfix] Fix multimodal configuration for Qwen3VL MOE model (#30670)
Max Hu committed 11 months ago
[Refactor] [3/N] Move tool parser tests and run on CPU (#30693)
Cyrus Leung committed 11 months ago
[Refactor] [2/N] Move tool parsers into the vLLM main directory (#30675)
Chauncey committed 11 months ago
Nicolò Lucchesi committed 11 months ago
[BugFix] Add embed_input_ids method to make QWenLMHeadModel a vllm model (#30674)
duke committed 11 months ago
[Model] Automatic conversion of TokenClassification model (#30666)
wang.yuqi committed 11 months ago
typing: Add type hints to TurnMetrics class in context.py (#30552)
ゆり committed 11 months ago
[New Model] BAGEL support (AR only) (#28439)
汪志鹏 committed 11 months ago
[XPU] fix Dockerfile.xpu, avoid wheel conflicts (#30662)
Kunshang Ji committed 11 months ago
[Feat] Refactor for `parallel_config` in `FusedMoEModularKernel` (#30282)
Wentao Ye committed 11 months ago
[Bugfix] Drop empty tool_calls lists to keep assistant replies in chat template (#30648)
Seokhyun An committed 11 months ago
[Bugfix] Fix deepseek_v32 tokenizer_mode (#30658)
Jee Jee Li committed 11 months ago
Shanshan Shen committed 11 months ago
additional protection for CVE-2025-62164 (#30649)
Wenqi Glantz committed 11 months ago