[Bugfix][LoRA] Guard None group members in expand_packed_lora (partial LoRA on Qwen3.5/3.6 GatedDeltaNet) (#47640)

Eilam committed 2 months ago

[Bugfix][Frontend] Return all choices from /inference/v1/generate when n > 1 (#52399)

Quentin Gallouédec committed 2 months ago

Add NemotronH_Omni_Reasoning_V3 as a supported Nemotron architecture (#52929)

Nave Assaf committed 2 months ago

[CI] Fix docs build (#52937)

Harry Mellor committed 2 months ago

[Doc] Fix group numbering in Case 3 of hybrid_kv_cache_manager.md (#52160)

QWERQWERQWE86 committed 2 months ago

[Bugfix] Restore model info caching for package backends (#52690)

hy123 committed 2 months ago

[Platform] Fill in the missing backend parameter for torch.compile (#51781)

wangxiyuan committed 2 months ago

[Frontend] Move api_server.py out openai folder (#52131)

wang.yuqi committed 2 months ago

fix: reject string schemas that mix pattern/format with length bounds (#49996)

Yufeng He committed 2 months ago

[XPU][CI] downgrade sentencepiece (#52904)

Qiming Zhang committed 2 months ago

[Bugfix] Redact api_key in non-default args log (#52523)

LuckyStep committed 2 months ago

[Pooling] Use semantic task validation errors (#52867)

Taneem Ibrahim committed 2 months ago

[Model][NVIDIA] Route DSA models to the CUDA non-compiled path (#52861)

Woosuk Kwon committed 2 months ago

[BugFix] Revert incorrect MM keep_on_cpu=True changes (#52881)

Nick Hill committed 2 months ago

[Bugfix][LoRA] Add embedding_modules for Qwen3.5 CausalLM (#48850)

Agoni-02 committed 2 months ago

[Bugfix][Frontend] Run the serve arg checks for `vllm launch` too (#52825)

Vineeth Sai Varikuntla committed 2 months ago

[ModelRunner v2] Enable MRV2 for pooling models by default (#48290)

Taneem Ibrahim committed 2 months ago

[Attention][MLA] FlashMLA sparse: DCP on the fp8_ds_mla mixed-batch path + MTP (#46514)

drakosha committed 2 months ago

[Model] Add GraniteSWA and GraniteMoeSWA via existing Granite (#52706)

Davis Wertheimer committed 2 months ago

[Attention] Vectorize sparse MLA mask loads (#52217)

Matthew Bonanni committed 2 months ago