[Rust Frontend] Simplify data-parallel size ownership (#52575)

Bugen Zhao committed 2 months ago

Harden DeepSeek V3.2 fused kernel grids (#52381)

yimdev committed 2 months ago

[Bugfix][ROCm] Fix a few int4/int8 quantization errors (#52112)

qli88 committed 2 months ago

[Bugfix] Accept logprobs=-1 in the Completion API (#46175)

Yufeng He committed 2 months ago

[Bugfix] Return 4xx for client-caused errors in /detokenize (#52622)

Rajath Pai committed 2 months ago

Replicated embedding and norm fusion for DSV3 flat model (#48484)

Jee Jee Li committed 2 months ago

[ROCm][CI] Move ROCm AITER quantization tests (#40938)

Andreas Karatzas committed 2 months ago

[Test] Add pause/resume E2E tests (#52144)

flb_ committed 2 months ago

[Build] Propagate vLLM version to Rust binaries (#52593)

Bugen Zhao committed 2 months ago

[XPU] update xpu-manager to v2.1.0 (#52569)

Yan Ma committed 2 months ago

[Cohere] Misc changes to cohere model definitions (#50156)

Komal Kumar Teru committed 2 months ago

[Bugfix] Handle DeepseekV4ForCausalLM in benchmark_moe get_model_params (#52044)

Do_it_now_! committed 2 months ago

[Rust Frontend] Fix GLM-5.2 chat template rendering parity (#51426)

Woosuk Kwon committed 2 months ago

[Bugfix] Fix DeepSeek V4 mHC broadcast buffer for weight sync (#52626)

ℍ𝕠𝕝𝕝𝕠𝕨 𝕄𝕒𝕟 committed 2 months ago

[XPU][UT] Fix OOM and skip graph case (#49287)

Qiming Zhang committed 2 months ago

[Bugfix][Quantization] Guard the MXFP8 FlashInfer path on FlashInfer availability (#52648)

Han committed 2 months ago

[Kernel][Perf] Support Qwen head ratios in fused GDN MTP (#52539)

kiroxu committed 2 months ago

[MoE] Refine FlashInfer one-sided All2All integration (#51924)

Bo Li committed 2 months ago

[Kimi-K3] support DCP partial prefix cache hit (#50493)

Summer Yang committed 2 months ago

[ROCm] gaurd on_gfx1250 call with rocm platform (#52625)

Kunshang Ji committed 2 months ago