[Bugfix] Fix Gemma4 startup crash with recent transformers multimodal processor (#44232)

Luciano Martins committed 5 months ago

[Bugfix] Sync block_size from EngineCore to frontend for hybrid Mamba… (#42967)

gruner committed 5 months ago

[Frontend] Consolidate dev entrypoints. (#44170)

wang.yuqi committed 5 months ago

Fix DFlash prefix cache corruption due to missing lookahead block (#42971)

Shreyas Kulkarni committed 5 months ago

[XPU][Mamba] Triton-based selective scan forward op for XPU (#43421)

Marceli Fylcek committed 5 months ago

[KV Offload] Add `on_schedule_end()` hook to separate step lifecycle from event draining (#44206)

Ronen Schaffer committed 5 months ago

[HARDWARE][POWER] Enable SHM communicator support for PowerPC (#43754)

Rukhaiya2004 committed 5 months ago

[Multimodal] Automatically select registered video loader for VLM (#44126)

Isotr0py committed 5 months ago

[Parser] Migrate `ResponsesParser` to unified `Parser` interface (#42977)

alberto committed 5 months ago

[EC Connector] Non blocking EC Connector lookup (#41627)

omerpaz95 committed 5 months ago

[Core][Refactor]: thread `scheduler_block_size` into KVCacheManager and KVCacheCoordinator (#44165)

Yifan Qiao committed 5 months ago

[Rust Frontend] add --enable-request-id-headers flag support. (#43883)

Maria Guevara committed 5 months ago

[kv_offload] Add `@override` decorators to subclass method implementations (#44177)

Ronen Schaffer committed 5 months ago

[CPU][Perf] Enable fused kernels for GDN's gated delta rules (#43534)

Fadi Arafeh committed 5 months ago

[Perf] use triton moe backend on hopper by default (#44220)

Jiangyun Zhu committed 5 months ago

[Refactor] Unify reasoning + tool-call parsing behind Parser.parse() (#44267)

Flora Feng committed 5 months ago

[Model Runner V2] Use actual batch max_seq_len for attn metadata (#43991)

zhrrr committed 5 months ago

[Model Runner V2] Support zeroing freshly allocated KV blocks for hybrid + fp8 KVCache (#43990)

zhrrr committed 5 months ago

[Bugfix] Convert Gemma4-MM ViT linear layers to vllm native impl (#43798)

Isotr0py committed 5 months ago

[MM][CG] Profile encoder CUDA graph pool memory (#41714)

JooHo Lee committed 5 months ago