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, 05:00 UTC
main
[BugFix] Partial revert of #29558 (DeepEP HT + PIECEWISE CG support) (#30910)
Lucas Wilkinson committed 11 months ago
[Bugfix] Fix Unicode issues in GLM-4 tool calling (#30920)
Chauncey committed 11 months ago
Andreas Karatzas committed 11 months ago
[Platform] Let EPD work with non-cuda platform (#30225)
wangxiyuan committed 11 months ago
[Fix][FlexAttention] return max logical block index to handle reused blocks (#30915)
Yifan Qiao committed 11 months ago
[CPU] Refactor CPU fused MOE (#30531)
Li, Jiang committed 11 months ago
fix: add warmup for audio preprocessing (#30706)
Nathan Price committed 11 months ago
[KV connector][LMCache] Only record the cuda event when there are request to store/load (#30814)
Yihua Cheng committed 11 months ago
[AMD][CI] fix lm eval ci arg (#30911)
Divakar Verma committed 11 months ago
[PERF] Qwen3-next. Add fp8 cutlass MoE tuned configs. `chmod -x *MI308X.json` (#29553)
Vadim Gimpelson committed 11 months ago
[Doc][CPU] Update CPU doc (#30765)
Li, Jiang committed 11 months ago
[refactor] Add prefix support to embed_tokens in DeepSeek MTP (#30788)
zzhxxx committed 11 months ago
[compile] Fix CI for test_gpt2_cache_hit (#30902)
Zhengxu Chen committed 11 months ago
[UX] Reduce DeepGEMM warmup log output to single progress bar (#30903)
Matthew Bonanni committed 11 months ago
[Quantization] Support Quark int4-fp8 w4a8 for MoE (#30071)
Bowen Bao committed 11 months ago
fused_moe_lora PDL improvements (#30716)
gnovack committed 11 months ago
[Chore] Remove v0 dead code for Qwen2.5-omni (#30883)
Isotr0py committed 11 months ago
[NIXL] Support P tensor-parallel-size > D tensor-parallel-size (#27274)
Nicolò Lucchesi committed 11 months ago
[ROCm][CI] Reduce Flakiness For test_async_scheduling Using ROCM_ATTN With FP32 (#30811)
Micah Williamson committed 11 months ago
[Metrics] Model FLOPs Utilization estimation (#30738)
SungMinCho committed 11 months ago