[Nixl][PD] Lease renewal TTL KV blocks on P (#41383)

Nicolò Lucchesi committed 6 months ago

[Model] Fix missing `maybe_prefix` (#42280)

Cyrus Leung committed 6 months ago

bugfix(flashinfer,dcp): remove kv_cache_layout for BatchDCPPrefillWrapper._new_tokens. (#38895)

Qiu committed 6 months ago

fix nixl side-channel host selection (#41806)

shaharmor98 committed 6 months ago

Document MolmoWeb hf_overrides (#42163)

Haoqing Wang committed 6 months ago

[CI/Build] Reduce LoRA model tests. (#42266)

Jee Jee Li committed 6 months ago

[Frontend] Return rendered prompt text in chat completion response (#42052)

汪志鹏 committed 6 months ago

[Misc] Replace mamba_type string literals with MambaAttentionBackendEnum (#41188)

wangxiyuan committed 6 months ago

[ROCm][Bugfix]: dynamically align BLOCK_DMODEL with Lv in MLA decode kernel (#41119)

vllmellm committed 6 months ago

[Bugfix] Gemma 4 chat template crash with missing tool name and tool id (#42188)

yzong-rh committed 6 months ago

add fused mhc_post_pre kernel (#41536)

gnovack committed 6 months ago

Fix Molmo2 image token metadata (#42162)

Haoqing Wang committed 6 months ago

[MoE] Move various experts classes to fused_moe/experts/ (#41979)

bnellnm committed 6 months ago

[Performance] Make safetensors checkpoint prefetch settings configurable (#41499)

Mohammad Miadh Angkad committed 6 months ago

[DSV4] Add PP support for deepseek-v4 (#41694)

Isotr0py committed 6 months ago

[Bugfix][Mamba] IMA in causal_conv1d kernel for long sequences (#41617)

Vensen committed 6 months ago

[KV Offload] Pass ReqContext to touch(), complete_load(), and complete_store() (#41366)

Ronen Schaffer committed 6 months ago

[CPU] Fix spec decode kernel signatures for synthetic mode compatibility (#41932)

Jonathan Mamou committed 6 months ago

[Bugfix] Fuse Qwen3.5 in_qkvz_proj forwarding with LoRA enabled (#37912)

Isotr0py committed 6 months ago

[ROCm] Cap Triton paged attention block size to fix ROCm shared memory OOM (#38502)

Andreas Karatzas committed 6 months ago