vllm-project/vllm

Pull Requests Commits

Merge branch 'main' into worktree-fix-cudagraph-flaky

tlrmchlsmth committed 38 days ago

Verified 6257a61c

[Kernel][Helion][1/N] Add Helion kernel for silu_and_mul_per_block_quant (#43994)

xiaohongchen1991 committed 38 days ago

Verified 9fde043f

[Bugfix] Preserve FP8 indexer WK pairs across incremental load_weights (#46168)

lcheng321 committed 38 days ago

Verified 24dd2aec

[macOS][CPU][Installation] Fix the broken installation of vllm 0.24.0 in macos + cpu (#47457)

WindChimeRan committed 38 days ago

Verified 3ee9eea9

Make the Transformers modeling backend as fast as native vLLM (#47187)

hmellor committed 38 days ago

Verified 5bce653e

[perf]Add fused Kimi image preprocessing (#47416)

Kevin-XiongC committed 38 days ago

Verified 5ad11172

[Perf] Cache `token_to_req_indices` for dsv4, 5x~6x kernel performance improvement (#47474)

yewentao256 committed 38 days ago

Verified f70caef4

[Bugfix][Spec Decode] Add missing draft_id_to_target_id to DSparkDeepseekV4ForCausalLM (#47429)

Laurent-Zhang committed 38 days ago

Verified 8d8ec383

[Refactor] Remove multiple dead code (#47329)

yewentao256 committed 38 days ago

Verified b1c6dba5

[Bugfix][Distributed] Delegate MNNVL allreduce one-shot selection (#47589)

jesco-absolut committed 38 days ago

Verified 598d5115

[Bugfix] Fix int32 overflow in triton_decode_attention page offsets (#47671)

ivanium committed 38 days ago

Verified 095adf1f

[CI] Skip test for checkpoint that was deleted (#47748)

hmellor committed 38 days ago

Verified 51ee564e

[Bugfix][Core] Fix num_output_placeholders underflow with async scheduling + spec decode (#46066)

Sunt-ing committed 38 days ago

Verified 373eb314

[Doc] Clarify fastokens availability (#45813)

LiJzd committed 38 days ago

Verified 641cb595

Revert "[Platform] Replace `torch.cuda.Event` with `torch.Event` (#47140)" (#47668)

jikunshang committed 38 days ago

Verified 07f9baf7

[Bugfix] [Gemma4] Fix Gemma4 MTP draft model layers ignoring quant_config (#47091)

ayush1399 committed 38 days ago

Verified 7a90eb98

[Rust Frontend] Cache metric handles for scheduler & request stats (#47444)

BugenZhao committed 38 days ago

Verified 8f4c69b2

attention: pass None for unused args in unified attention TD path (#43597)

afierka-intel committed 38 days ago

Verified 8b79971b

[CI] Use TTY for AMD CI tests for colored buildkite logs (#47730)

njhill committed 38 days ago

Verified f676808b

[fix][run_batch]: respect proxy env vars when downloading media URLs (#47697)

mayuyuace committed 38 days ago

Verified 98e4726a

[ROCm][AITER] Directly Implement AITER Custom All-reduce in CudaCommunicator (#46065)

BadrBasowid committed 38 days ago

Verified 740f379f

[Bugfix] Return HTTP 422 for unprocessable image URLs instead of 500 (#47165)

akinsella committed 38 days ago

Verified 40cc2e83

fix(security): block request-level GPU video backend selection withou… (#47259)

jperezdealgaba committed 38 days ago

Verified ba221520

[bugfix] fix MOSS-Audio deepstack_input_embeds initialization in PP (#47607)

yma11 committed 38 days ago

Verified 90ce3a09

[XPU][Bugfix] Do not transpose weight_scale_inv at load time (#47116)

majian4work committed 38 days ago

Verified 26c754d8

[Doc] docs: fix note formatting for pooling models (#47701)

llsj14 committed 38 days ago

Verified 3d7f357e

[XPU] Route mm_prefix models to Triton attention backend (#47688)

zhenwei-intel committed 38 days ago

Verified 736f1a59

[CI/Build] Fix pre-commit check (#47695)

bigPYJ1151 committed 38 days ago

Verified 344609ab

[Bugfix] Recycle post-final-norm hidden in GLM MTP (single norm) (#47448)

zhou9402 committed 38 days ago

Verified d039c171

[XPU][CI]Add agent tags for Basic Models Tests (Initialization) in Intel GPU CI (#47675)

zxd1997066 committed 38 days ago

Verified cdab2831

Older