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 6, 14:30 UTC
main
[compile] Enable mega aot artifact for torch 2.12+. (#37198)
Zhengxu Chen committed 8 months ago
[Bugfix][MultiConnector] Fix MultiConnector for SupportsHMA sub-connectors (#36549)
zhanqiuhu committed 8 months ago
[BUGFIX][Mamba] Use uint64 for address in KVBlockZeroer (#37197)
Kunshang Ji committed 8 months ago
[Misc] Add `float16` to `CacheDType` (#37199)
Matthew Bonanni committed 8 months ago
[CI][BugFix][MORI][AMD] Add transfer_id to kv transfer params for test (#37213)
rasmith committed 8 months ago
[Refactor] Consolidate GPT-OSS reasoning parser tests (#36915)
Flora Feng committed 8 months ago
[torch.compile][BE] Modify cudagraph callable to check for is_forward_context_set (#36288)
Lucas Kabela committed 8 months ago
[Bugfix] Fix render server crash for quantized models on CPU-only hosts (#37215)
Sage committed 8 months ago
[PD][Nixl] Add support for hybrid SSM-FA models (#36687)
Nicolò Lucchesi committed 8 months ago
[MTP][Sparse MLA] Take advantage of native MTP support in indexer when possible (#36982)
Matthew Bonanni committed 8 months ago
Add simple granite4 tool parser (#36827)
Max de Bayser committed 8 months ago
[Bugfix] Disable cross-layer KV cache for MLA attention backends (#37090)
haosdent committed 8 months ago
[Bugfix] Make siglip/clip compatible with transformers v5 (#37200)
Raushan Turganbay committed 8 months ago
[Bugfix] Add error handling for FINISHED_ERROR in OpenAIServing (#37148)
Chauncey committed 8 months ago
[Bugfix] Relax TRTLLM KV cache contiguity assertion for cross-layer layout (#34158)
Itay Etelis committed 8 months ago
[Compile] Fix compile warning `st256_cs` in `cuda_vec_utils.cuh` (#36693)
Wentao Ye committed 8 months ago
[BUGFIX]fix CUDA OOM ERROR : invalid argument at cumem_allocator.cpp:119 (#35594)
xjx committed 8 months ago
[Bugfix] Add safety check and fallback for null scaling factor (#36106)
Yuanheng Zhao committed 8 months ago
[Compile] Fix compile warning in `moe_permute` (#36529)
Wentao Ye committed 8 months ago
[Bugfix] accept redacted thinking blocks in Anthropic messages (#36992)
Benjamin Bartels committed 8 months ago