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 12, 17:15 UTC
main
[CI/Build] Add terratorch for AMD (#29205)
Ryan Rock committed 12 months ago
[Model][Qwen3VL] Tune Triton w8a8 block fp8 kernel for L40s (#29217)
Lukas Geiger committed 12 months ago
[CI] Bug: Fix triton import issue (#29202)
Wentao Ye committed 12 months ago
[Rocm][CI] Fix DeekSeek V2-Lite Accuracy CI (#29135)
Charlie Fu committed 12 months ago
[docs] Fix cudagraph mode config (#29170)
Angela Yi committed 12 months ago
[BugFix] Fix Eagle `IndexError: list index out of range` for even `num_speculative_tokens` (#29102)
Lucas Wilkinson committed 12 months ago
[Model Runner V2] Optimize Gumbel Sampling Kernel (#29210)
Woosuk Kwon committed 12 months ago
[KV Connector] Fix async connector prefix cache metrics (#28585)
Mark McLoughlin committed 12 months ago
[BugFix] EPLB + B200 + DeepGEMM : Handle column-major scales tensor (#29162)
Varun Sundar Rabindranath committed 12 months ago
Default model load/config/tokenizer to `mistral` format if relevant files exist (#28659)
Julien Denize committed 12 months ago
[BugFix] Fix missing symbol triggering FA2 fallback on Hopper (#29107)
Lucas Wilkinson committed 12 months ago
[log] add weights loading time log to sharded_state loader (#28628)
Ning Xie committed 12 months ago
Lucas Wilkinson committed 12 months ago
[Chore] Fix pre-commit error after #25266 (#29190)
Woosuk Kwon committed 12 months ago
[Deprecation] Deprecate `seed=None` (#29185)
Cyrus Leung committed 12 months ago
[ROCm][CI] Fix config/test_config_generation.py (#29142)
Charlie Fu committed 12 months ago
[NIXL] Fix after virtual block_size for host_buffer with heter kv_layout (#29122)
Chendi.Xue committed 12 months ago
Add TRTLLM MoE NVFP4 kernel to CompressedTensorsW4A4MoeMethod (#28892)
Mingyuan Ma committed 12 months ago
rasmith committed 12 months ago
[Log] Optimize startup log (#28948)
Wentao Ye committed 12 months ago