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 7, 16:30 UTC
main
[ROCm] Add dynamic mxfp4 quantization for DeepSeek V2 projection layers (#34157)
Douglas Lehr committed 8 months ago
[Refactor] Remove dead or duplicate func utils or variables (#35318)
Wentao Ye committed 8 months ago
[Model] Add nvidia/llama-nemotron-embed-vl-1b-v2 multimodal embedding model (#35297)
Jakub Zakrzewski committed 8 months ago
[Bug] Fix missing <think> tag after tool call in MiniMax 2.1 (#35352)
stingoChen committed 8 months ago
[Misc] Standardize handling of `mm_processor_kwargs.size` (#35284)
Cyrus Leung committed 8 months ago
[Bugfix] Fix uint32 overflow in Mamba selective scan state pointer arithmetic (#35275)
Asaf Gardin committed 8 months ago
[Bugfix] fix device_name for routing replay (#34336)
Li-Yongwen committed 8 months ago
[Bugfix] Fix Qwen2.5-Omni and Qwen3-Omni mixed-modality embed regression (#35368)
Yueqian Lin committed 8 months ago
Harry Mellor committed 8 months ago
[XPU] use fixed UMD version in dockerfile.xpu (#35392)
Kunshang Ji committed 8 months ago
[Bugfix] [Qwen3.5]Fix Qwen3.5 FP8 quantization: tuple shard_id weight loading (#35289)
HZY committed 8 months ago
[Hardware][Powerpc]Enable prefix caching and chunked prefill for ppc64le (#35081)
Akash kaothalkar committed 8 months ago
[Benchmarks] Plot benchmark timeline and requests statistics (#35220)
Sophie du Couédic committed 8 months ago
Jiangyun Zhu committed 8 months ago
[Test] Add tests for n parameter in chat completions API (#35283)
Krish Gupta committed 8 months ago
[Bugfix][Hardware][AMD] Gate FP4 ops on gfx950 to prevent MI300X crash (#35250)
Kevin McKay committed 8 months ago
Gregory Shtrasberg committed 8 months ago
[XPU][8/N] Fix kernel bugs in XPU LoRA and MOE LORA (#34115)
Chaojun Zhang committed 8 months ago
Ofir Zafrir committed 8 months ago
[Benchmark] Simplify SLA scan (#35306)
Cyrus Leung committed 8 months ago