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, 03:45 UTC
main
[Frontend][Bug] allow tool calls in analysis channel (#28139)
Marko Rosenmueller committed 11 months ago
[Bugfix] Add validation for tool requests when tool_parser is unavailable (#30613)
lif committed 11 months ago
[Quantization] enable compressed-tensors marlin support for turing (2) (#31008)
Jinzhen Lin committed 11 months ago
[Bugfix] Fix incorrect tiles creation for mm prefix triton attention (#30974)
Isotr0py committed 11 months ago
Add hidden dimension validation for multimodal embedding inputs (#30968)
Wenqi Glantz committed 11 months ago
[Doc][CPU] Fix index link for CPU regular release wheels (#31015)
Li, Jiang committed 11 months ago
[Quantization] enable compressed-tensors marlin support for turing (#31000)
Jinzhen Lin committed 11 months ago
[BugFix] Fix logprobs with spec decode and modified logits (#30846)
Nick Hill committed 11 months ago
[ROCm][CI][Bugfix] Multi-Modal Model Support Fixes and Attention Backend Improvements (#30270)
Andreas Karatzas committed 11 months ago
[Bugfix] Fix tool_choice="none" being ignored by GPT-OSS/harmony models (#30867)
PlatinumGod committed 11 months ago
[BugFix] Handle errors when preprocessing added requests (#30895)
Nick Hill committed 11 months ago
[Cleanup] Refactor FlashInferMetadataBuilder (#29128)
Benjamin Chislett committed 11 months ago
[NIXL][BUG FIX] Fix both failing issue and accuracy issue with nixl + host_buffer on CUDA (#30419)
Chendi.Xue committed 11 months ago
Check for truthy `rope_parameters` not the existence of it (#30983)
Harry Mellor committed 11 months ago
[BugFix] Fix spec decode + structured outputs + preemption edge case (#30916)
Nick Hill committed 11 months ago
Strengthen input validation and tests for 'parse_raw_prompts’. (#30652)
Kayvan Mivehnejad committed 11 months ago
[BugFix] Spec decode with VLLM_ENABLE_V1_MULTIPROCESSING=0 (#30319)
Chen Zhang committed 11 months ago
Remove all2all backend envvar (#30363)
Elizabeth Thomas committed 11 months ago
[Bug] Fix compressed tensor not using deepgemm (#30820)
Wentao Ye committed 11 months ago
[Bugfix] Remove `tile_size=64` for mm_prefix triton attention (#30973)
Isotr0py committed 11 months ago