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 14, 05:30 UTC
main
[BUGFIX ] fix undefined silu_and_mul_nvfp4_quant (#23929)
yzds committed about 1 year ago
[CI] Add `aiter` to matching list of issue auto labeller for `rocm` tag (#23942)
vllmellm committed about 1 year ago
[BugFix] Async scheduling and PP compatibility with DP (#23770)
Nick Hill committed about 1 year ago
[Models] Use in-place adds in Idefics2Vision (#23932)
Lukas Geiger committed about 1 year ago
[MODEL] `Apertus` and `XIELU` (#23068)
EduardDurech committed about 1 year ago
Adds `json_count_leaves` utility function (#23899)
Adit Chawdhary committed about 1 year ago
Update PyTorch to 2.8.0 (#20358)
Huy Do committed about 1 year ago
[Multimodal] Consolidate mm inputs into MultiModalFeatureSpec (#23779)
Flora Feng committed about 1 year ago
[Performance] V1 Classify Models E2E Performance Optimization (#23541)
wang.yuqi committed about 1 year ago
[CPU] Enable data parallel for CPU backend (#23903)
Li, Jiang committed about 1 year ago
[V0 Deprecation] Remove pooling model support in V0 (#23434)
Maximilien de Bayser committed about 1 year ago
Better errors for Transformers backend missing features (#23759)
Harry Mellor committed about 1 year ago
[Misc] Fix warnings for mistral model (#23552)
Jiangyun Zhu committed about 1 year ago
Chenheli Hua committed about 1 year ago
[CI/Build] Clean up LoRA test (#23890)
Jee Jee Li committed about 1 year ago
Fix(async): Add support for truncate_prompt_tokens in AsyncLLM (#23800)
Raghavan committed about 1 year ago
[Platform] import activation_quant_fusion for CUDA only (#23882)
wangxiyuan committed about 1 year ago
[BugFix][AMD][Deepseek] fix a dtype mismatch error for deepseek running on AMD (#23864)
Jinghui Zhang committed about 1 year ago
[Bugfix] Use `ReplicatedLinear` for SequenceClassification head (#23836)
Isotr0py committed about 1 year ago
[tests] Improve speed and reliability of test_transcription_api_correctness (#23854)
Russell Bryant committed about 1 year ago