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, 07:30 UTC
main
[Fix]Load kv-cache dtype from hf_quant_config.json automatically (fix for reverted PR) (#30785)
danielafrimi committed 11 months ago
[compile] Ignore VLLM_FORCE_AOT_LOAD from cache factors (#30809)
Zhengxu Chen committed 11 months ago
[compile] Disable aot when eager backend is used. (#30810)
Zhengxu Chen committed 11 months ago
[UX] Make `vllm bench serve` discover model by default and use --input-len (#30816)
Michael Goin committed 11 months ago
[Kernels][FI] Skip trtllm attention when num_kv_heads=1 (#30842)
Ye (Charlotte) Qi committed 11 months ago
[Doc][ResponsesAPI] add documentation (#30840)
Andrew Xia committed 11 months ago
[NIXL][Bugfix] Fix NIXL/RDMA registration failure over CuMemAllocator (#29569)
Sheng Lin committed 11 months ago
CustomOp: grouped topk (#29575)
Xinyu Chen committed 11 months ago
[Mamba] Removed disable cascade attn in MambaModelConfig (#30712)
Asaf Joseph Gardin committed 11 months ago
[Bugfix][Frontend] Prevent IndexError in MiniMax M2 tool parser during streaming extraction (#30555)
Robin committed 11 months ago
[XPU] fix broken fp8 online quantization for XPU platform (#30831)
Yan Ma committed 11 months ago
[Bugfix][CPU] Fix CPU backend ROPE dispatch for VL models (#30829)
Li, Jiang committed 11 months ago
Update note comment for flashinfer attention warmup (#30711)
Michael Goin committed 11 months ago
bump up compressed tensors version to 0.13.0 (#30799)
shanjiaz committed 11 months ago
[CI/Build] Fix compatibility between #30244 and #30396 (#30787)
Cyrus Leung committed 11 months ago
[CPU] Add action to automatically label CPU related PRs (#30678)
Fadi Arafeh committed 11 months ago
Update model-hosting-container-standards to 0.1.10 (#30815)
Michael Goin committed 11 months ago
Fix nemotron_nas intermediate_size computation (#30795)
Grzegorz K. Karch committed 11 months ago
[docker] Allow kv_connectors install to fail on arm64 (#30806)
Amr Mahdi committed 11 months ago
[MM] Pass FA version in ViT Attn (#30756)
Nicolò Lucchesi committed 11 months ago