Commits vllm-project/vllm

Remove unused disable_fallback field

zhuohan123 committed 6 hours ago

Verified d5474912

[MTP][Misc] Clean up dead code (#36507)

MatthewBonanni committed 8 hours ago

Verified 4e571ce6

[ROCm][CI] Prep Tests For Change To ROCM_ATTN As New Default Backend On ROCm (#36025)

micah-wil committed 8 hours ago

Verified 4ff9b045

[BE] Rename `should_torch_compile_mm_vit` to `should_torch_compile_mm_encoder` (#36281)

Lucaskabela committed 8 hours ago

Verified 3fd03f1e

[Model Runner V2] Use NamedTuple for `execute_model_state` (#35930)

WoosukKwon committed 8 hours ago

Verified 10a5f4d5

[Docs] Remove the reo beacon (#36528)

simon-mo committed 8 hours ago

Verified fe0c085c

[Misc] Refactored 5 duplicate helper functions that were copied-pasted across multiple parsers (#36436)

taneem-ibrahim committed 8 hours ago

Verified 8d6b3d5d

[torch.compile] Rename `compile_ranges_split_points` to `compile_ranges_endpoints` (#36027)

Copilot committed 9 hours ago

Verified 4b87ffbe

Fix/resupport nongated fused moe triton (#36412)

shaunkotek committed 9 hours ago

Verified fa028207

[Docs] Expand --allowed-media-domains security guidance with threat details (#36506)

russellb committed 9 hours ago

Verified d460a18f

[Model Runner V2] Add dummy profile_cudagraph_memory API (#36520)

WoosukKwon committed 9 hours ago

Verified 6e956d9e

[ROCm][CI] Fix logprob divergence for TitanML/tiny-mixtral under AITER rms_norm (#36101)

AndreasKaratzas committed 10 hours ago

Verified 1e0f917b

[ROCm][CI] Fix ROCm attention backend validation for head sizes, block sizes, and compute capability checks (#36292)

AndreasKaratzas committed 10 hours ago

Verified c174d54f

[Misc] fix typo: dependant -> dependent (2 lines change) (#36511)

SoluMilken committed 10 hours ago

Verified 55d27cca

[Attention][Perf][Kernel] Replace torch.cat with vectorized CUDA kernel MLA query concat - DeepSeek-V3.2 (#34917)

LopezCastroRoberto committed 10 hours ago

Verified 580864d8

[Attention][Perf] Optimize cp_gather_and_upconvert_fp8_kv_cache - DeepSeek-v3.2 (#35290)

LopezCastroRoberto committed 10 hours ago

Verified 2b28b9b2

[ROCM] Optimize the fused_topk_bias to use aiter instead of fallback torch ops. (#36253)

benenzhu committed 10 hours ago

70485a11

[CI] Fix edge case that could lead to broken docs builds on main (#36515)

hmellor committed 11 hours ago

Verified 74a9f54c

[Bugfix] Clear stale CG keys after memory profiling (#36416)

MatthewBonanni committed 11 hours ago

Verified 00c4cb56

[Refactor] Simplify `chat_completion_full_generator` for tool parsers (#35634)

yewentao256 committed 11 hours ago

Verified 941e52c2

[Bug] Fix pooling model benchmark script (#36300)

yewentao256 committed 11 hours ago

Verified be292b7c

Reapply [Attention] Refactor `check_and_update_config` (#35122)

MatthewBonanni committed 12 hours ago

Verified 77a73458

Support online use_audio_in_video (#36319)

gty111 committed 12 hours ago

Verified 5578f2a4

[Frontend] Move warmup into Renderer (#36482)

DarkLight1337 committed 14 hours ago

Verified 3ec21150

[MM Encoder] Default to use TORCH_SDPA backend for ViT on Volta/Turing GPU (#36472)

Isotr0py committed 16 hours ago

Verified b0906d8b

[ci] Bound openai dependency to 2.24.0 (#36471)

khluu committed 16 hours ago

Verified aaf5fa9a

[Deprecation][1/2] Remove items deprecated in v0.18 (#36470)

DarkLight1337 committed 16 hours ago

Verified f96c3ab0

[Kernel] Add fused_sigmoid_gating_delta_rule_update kernel for Qwen3 Next (#35777)

xyang16 committed 20 hours ago

Verified dc6b5784

[XPU] Add test script of PD disaggregation (#36434)

zhenwei-intel committed 21 hours ago

Verified 1bc9c77f

[Frontend] Add Support for MM Encoder/Decoder Beam Search (Online Transcriptions) (#36160)

alex-jw-brooks committed 21 hours ago

Verified 65a4da15