[Rust Frontend][gRPC] Add LoRA lifecycle control (#52840)

Connor Carpenter committed 2 months ago

Exclude the cpu backend from vLLM's active-Triton-driver count (#53530) (#53530)

Minjang Kim committed 2 months ago

[Model][MoE] DeepSeek-V4: add opt-in FlashInfer moe_ep expert backend (#49636)

Md Saidul Hoque Anik committed 2 months ago

[Bugfix][Frontend] Redact hf_token in the non-default args log (#53625)

Moe Huzaifa committed 2 months ago

[Docs] Fix docstring continuation indentation in `routed_experts.py` (#53604)

Harry Mellor committed 2 months ago

[Bugfix][NIXL] Fix Mamba prefill truncation ordering (#53523)

Ziming Huang committed 2 months ago

[XPU] Fix sparse-MLA metadata sync (#52066)

Libin Tang committed 2 months ago

[Bugfix][LoRA] Restore tower and connector LoRA support for LFM2-VL (#53519)

Zupeng Wang committed 2 months ago

[CI] Increase Entrypoints Unit timeout after launcher suite growth (#53618)

Kevin H. Luu committed 2 months ago

[Kernel][Perf] Enable fused QK-norm + partial MRoPE + gate for Qwen3.6 (#52676)

kiroxu committed 2 months ago

[Bugfix] Keep grid dims for XD-RoPE models on a prefix-cache hit (#53456)

drakosha committed 2 months ago

[Bugfix] BailingMoeV3 KDA: skip absent metadata during CUDA graph profiling (#53593)

Nick Hill committed 2 months ago

[CI] Report torch-nightly results to PyTorch CRCR (#51830)

Andrey Talman committed 2 months ago

[Model] Remove ten deprecated model architectures (#53608)

Harry Mellor committed 2 months ago

[CI][The Rock] Increase flex attention abs tol (#53646)

rasmith committed 2 months ago

[Model Runner V2] batch-sharded sample (#50465)

Giancarlo Delfin committed 2 months ago

fix(security): enforce VLLM_MAX_AUDIO_CLIP_FILESIZE_MB on all audio paths (#53561)

Juan Pérez de Algaba committed 2 months ago

speculative decoding under tensor parallelism (TP>1) , workspace creation select max hidden dim of target and draft model (#52193)

Khushali Desai committed 2 months ago

[Refactor] Refactor batch invariance folder (#53619)

Wentao Ye committed 2 months ago

[Attention][Spec Decode] Support varlen trtllm-gen decode for adaptive verification (#52157)

Guan-Ming Chiu committed 2 months ago