[Kernel][DSV4.1] Fuse MXFP8 wo_b GEMM with sequence-parallel reduce-scatter (#57428)

Canlin Guo committed about 1 month ago

[Rust Frontend] Separate multimodal instrumentation from request timing (#58084)

Bugen Zhao committed about 1 month ago

[Kernel] Add opt-in load-time MXFP4 dequantization (#50814)

yinfengLiu committed about 1 month ago

[ROCm][Perf] Avoid extra reshape kernel in Qwen GDN output norm (#47842)

Mikko Tukiainen committed about 1 month ago

[Rust Frontend] Build full-output grammars from initialized reasoning parsers (#57340)

Bugen Zhao committed about 1 month ago

[Bugfix][Qwen2.5-VL] Honor video fps for temporal M-RoPE (#47736)

Ting SUN committed about 1 month ago

[Bugfix][Frontend] Validate mixed prompt embedding mask lengths (#57006)

shaohuaxi committed about 1 month ago

[Security] Reject min_tokens that exceeds the filled max_tokens default (#57731)

Juan Pérez de Algaba committed about 1 month ago

[Rust Frontend] Introduce parser-owned output grammar interfaces (#55269)

Bugen Zhao committed about 1 month ago

[Frontend][Rust] Add mm-processor benchmark for Rust frontend (#51922)

karthik committed about 1 month ago

[Spec Decode] Enable async scheduling for DFlash (#58065)

Jeff (Junze) Ma committed about 1 month ago

[ROCm][Bugfix] Explicitly reject FSE=1 with DPA+ETP deployment for DeepSeek-V4 (#57919)

Shanshan Shen committed about 1 month ago

[RL][Sleep] Retain frozen weights across level-2 sleep (#57891)

aoshen02 committed about 1 month ago

[Bugfix][Spec Decode] Cap DFlash/DSpark profiling query batch (#56448)

Yicong Wang committed about 1 month ago

[Rust Frontend] Add MiMo V2.5 parser support (#57933)

Bugen Zhao committed about 1 month ago

[Fast Start] Cache the MTP draft model in a separate daemon group (#57312)

siyu committed about 1 month ago

[Docker] Expose bundled vllm-rs on PATH (#57606)

Alec committed about 1 month ago

[XPU] Fix Nemotron FP8 LM-eval config: drop CUDA-only moe_backend and wire to new Buildkite job (#49685)

Chaojun Zhang committed about 1 month ago

[CI] Add pre-commit check that new tests are tethered to Buildkite jobs (#54867)

Turner Jabbour committed about 1 month ago

[Bugfix][GDN] Fix stateless first-chunk classification (#51565)

cherry77-cloud committed about 1 month ago