[Bugfix][EPD][Model Runner V2] Skip gather mm embeddings for encoder only instance (#51222)

Tianyu Guo committed 3 months ago

[Feat] Support thinking_token_budget in Model Runner V2 (#46727)

Chauncey committed 3 months ago

[CI] Re-enable FI autotune in GSM8K config for Qwen3.5-35B-A3B (#51293)

Artem Perevedentsev committed 3 months ago

fix pre-commit broken (#51341)

Kunshang Ji committed 3 months ago

[Kernel] Support Nvfp4 Cutedsl Moe Swiglu-oai and Relu2(non-gated) Activation (#47106)

Bi Tiekai committed 3 months ago

[V1] Copy NaN-in-logits counts to host asynchronously (#51304)

Nick Hill committed 3 months ago

[CI] Exclude KV-connector subtree from broad source dependencies (#51046)

Nicolò Lucchesi committed 3 months ago

[Quantization] Share online weight scales across TP (#49764)

Matej Sirovatka committed 3 months ago

[Model Runner V2] Fix -1 placeholder draft token ids in rejection sam… (#50939)

Giancarlo Delfin committed 3 months ago

docs(governance): refresh committers list, add TSC note, update project leads (#51300)

Simon Mo committed 3 months ago

[ModelRunner v2] Enable decoder token-wise pooling (#50931)

Taneem Ibrahim committed 3 months ago

fix: resolve silent request skipping in PRIORITY scheduling (#49206)

Tejas committed 3 months ago

[Misc] Upgrade fastsafetensors version, fix metadata is null (#50827)

rongfu.leng committed 3 months ago

[ModelRunner V2] Minor indexing optimizations (#51210)

Nick Hill committed 3 months ago

[Bugfix] Fix packed KV block zeroing stride (#50276)

wangxian001 committed 3 months ago

[Mypy Fix] Mypy fix for "vllm/model_executor/models/[aA][bB]" (#48977)

Wentao Ye committed 3 months ago

[ROCm][MLA] Use asm decode for non-divisor small head counts (#50578)

vanshbhatia-amd committed 3 months ago

[Frontend] Watch frontend processes during engine startup (#43417)

Bugen Zhao committed 3 months ago

[ROCm][CI] Update AITER AR+RMS e2e fusion counts for final-norm coverage (#51273)

Divakar Verma committed 3 months ago

[Bugfix] Keep mamba align prefill chunks block-aligned past last_cache_position (#51113)

Yifan Qiao committed 3 months ago