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 - pass 'max_num_tokens_padded' into 'moe_lora_align_block_size' (#27311)
gnovack committed about 1 year ago
[Log] Add Warning for `LLM(data_parallel_size=k)` single-process DP Usage (#27282)
Wentao Ye committed about 1 year ago
[Bugfix][CPU] Disable dual stream execution for experts on CPU (#27320)
Li, Jiang committed about 1 year ago
[Frontend][3/N] Improve all pooling task | Support binary embedding response (#27066)
wang.yuqi committed about 1 year ago
fixed reasoning streaming with tool_choice="required" (#24108)
ExtReMLapin committed about 1 year ago
Remove last `level` references not removed in #26355 (#27260)
Harry Mellor committed about 1 year ago
Update release pipeline for PyTorch 2.9.0 (#27303)
Huy Do committed about 1 year ago
[1/N][Platform] Cleanup useless function (#26982)
wangxiyuan committed about 1 year ago
[torch.compile] Enable silu_mul_fp8_quant fusion without custom ops enabled (#27146)
Jiangyun Zhu committed about 1 year ago
[Benchmark] Add plot utility for parameter sweep (#27168)
Cyrus Leung committed about 1 year ago
[CI] Nixl integration tests DP-EP (#27199)
Nicolò Lucchesi committed about 1 year ago
[DOC] [ROCm] Add ROCm quickstart guide (#26505)
vllmellm committed about 1 year ago
[Deepseek v3.2] Remove extra logics in indexer (#26465)
Lain committed about 1 year ago
[P/D] KVConnector for decode benchmarking (#25986)
Tyler Michael Smith committed about 1 year ago
[Bugfix] skip cuda graph for drafter when running with eager (#26821)
Benjamin Chislett committed about 1 year ago
Updated xgrammar backend to not deny supported string formats (#27253)
ExtReMLapin committed about 1 year ago
Alexander Matveev committed about 1 year ago
Update PyTorch to 2.9.0+cu129 (#24994)
Huy Do committed about 1 year ago
[Bugfix] fixes the decoding metadata of dense mla's fp8 kvcache. (#27144)
Tao He committed about 1 year ago
[V0 Deprecation] Remove V0 executors (#27142)
Nick Hill committed about 1 year ago