Commit Graph

Select branches

Hide Pull Requests

main

patches

sm86

Mono Color

• 0eebd74842 [Model][Gemma3] Simplify image input validation (#18710) Lukas Geiger 2025-05-27 04:13:37 +01:00
• 27bebcd897 Convert examples to ruff-format (#18400) Harry Mellor 2025-05-26 17:57:54 +01:00
• e7523c2e03 [V1][Sampler] Improve performance of FlashInfer sampling by sampling logits instead of probs (#18608) Lukas Geiger 2025-05-26 16:49:36 +01:00
• a869baca73 [Bugfix] Fix Llama GGUF initialization (#18717) Cyrus Leung 2025-05-26 22:49:22 +08:00
• 82e2339b06 [Doc] Move examples and further reorganize user guide (#18666) Cyrus Leung 2025-05-26 22:38:04 +08:00
• 9553fdb41e [Doc] Improve API docs (#18713) Cyrus Leung 2025-05-26 22:33:34 +08:00
• 243eb9199f [Bugfix]: handle hf-xet CAS error when loading Qwen3 weights in vLLM (#18701) dylan 2025-05-26 22:10:56 +08:00
• 0665e29998 [Misc] add AutoGen integration (#18712) 3 people 2025-05-26 21:56:18 +08:00
• e76be06550 [Hardware][Intel-Gaudi] [CI/Build] Add tensor parallel size = 2 test to HPU CI (#18709) Łukasz Durejko 2025-05-26 14:26:07 +02:00
• 0877750029 [CI/Build] Split pooling and generation extended language models tests in CI (#18705) Isotr0py 2025-05-26 19:00:08 +08:00
• 6d68030f1c [Model] Add support for YARN in NemotronNAS models (#18427) Naveassaf 2025-05-26 13:31:49 +03:00
• 5a2c76cbe1 [CI] fix dump_input for str type (#18697) Ning Xie 2025-05-26 18:23:35 +08:00
• 38b13dfe78 [CI/Build] Replace math.isclose with pytest.approx (#18703) Cyrus Leung 2025-05-26 17:05:17 +08:00
• 61a45e7a72 [Bugfix] Fix Mistral-format models with sliding window (#18693) Cyrus Leung 2025-05-26 16:44:04 +08:00
• 65523a0995 [Doc] Fix issue template format (#18699) Cyrus Leung 2025-05-26 15:45:39 +08:00
• 4b7740a105 [GH] Add issue template for reporting CI failures (#18696) Cyrus Leung 2025-05-26 15:42:04 +08:00
• 4ea62c0ea0 [CI] add missing argument (#18694) Ning Xie 2025-05-26 15:22:04 +08:00
• 561b77a0d6 [Bugfix] Fix the lm_head in gpt_bigcode in lora mode (#6357) Maximilien de Bayser 2025-05-26 03:52:25 -03:00
• abd4030d94 refactor: simplify request handler, use positive condition check for handler assignment (#18690) CYJiang 2025-05-26 14:32:28 +08:00
• 8820821b59 [Misc] Fixed the abnormally high TTFT issue in the PD disaggregation example (#18644) AlexZhaoandzhaohaiyuan 2025-05-26 13:51:27 +08:00
• fba0642704 [CI/Build][Doc] Update gte-Qwen2-1.5B-instruct usage (#18683) Cyrus LeungandIsotr0py 2025-05-26 11:27:50 +08:00
• 6071e989df [Core][Multimodal] Convert PIL Image to array without data copy when hashing (#18682) Lukas Geiger 2025-05-25 18:33:35 +01:00
• 57fd13a707 [Bugfix] Fix profiling dummy data for Pixtral (#18677) Cyrus Leung 2025-05-25 22:05:30 +08:00
• 3a886bd58c [Misc] small improve (#18680) Reidandreidliu41 2025-05-25 21:05:38 +08:00
• 35be8fad62 [CI/build] fix no regex (#18676) Reidandreidliu41 2025-05-25 18:10:51 +08:00
• f2faac745d [Bugfix] Fix cpu usage and cache hit stats reporting on cpu environment (#18674) Yuqi ZhangandCyrus Leung 2025-05-25 02:36:06 -07:00
• 279f854519 [doc] improve readability (#18675) Reidandreidliu41 2025-05-25 16:40:31 +08:00
• 624b77a2b3 [doc] fix broken links (#18671) Reidandreidliu41 2025-05-25 16:36:33 +08:00
• 503f8487c2 [Misc] Reduce logs on startup (#18649) Cyrus Leung 2025-05-25 14:03:53 +08:00
• 44073a7ac3 [BUGFIX] catch subclass first for try...except (#18672) Ning Xie 2025-05-25 13:34:24 +08:00
• 63934543a0 Speed up the kernels/quantization/ tests (#18669) Michael Goin 2025-05-25 01:02:59 -04:00
• 75f81750f3 [VLM] Initialize video input support for InternVL models (#18499) Isotr0pyandCyrus Leung 2025-05-25 12:51:25 +08:00
• 6ab681bcbe [Misc][ModelScope] Change to use runtime VLLM_USE_MODELSCOPE (#18655) Mengqing CaoandIsotr0py 2025-05-25 12:51:21 +08:00
• cebc22f3b6 [Misc]Replace cuda hard code with current_platform in Ray (#14668) Chenguang Li 2025-05-25 11:26:31 +08:00
• 6c6dcd8611 [MISC] correct signature for LoaderFunction (#18670) Ning Xie 2025-05-25 11:17:47 +08:00
• 7891fdf0c6 [V1] Fix _pickle.PicklingError: Can't pickle <class 'transformers_modules.deepseek-ai.DeepSeek-V2-Lite... (#18640) Seiji Eicher 2025-05-24 20:07:20 -07:00
• 6825d9a998 [BugFix][Spec Decode] Improve Prefix Caching Logic in Speculative Decoding (#18668) Woosuk Kwon 2025-05-24 17:33:46 -07:00
• b554ab736e [CI/Build] fix permission denied issue (#18645) Reidandreidliu41 2025-05-25 00:09:10 +08:00
• 9ea7f1abf3 fix(regression): clone from reference items (#18662) Aaron Pham 2025-05-24 11:25:20 -04:00
• 2807271c86 [CI] enforce import regex instead of re (#18665) Aaron Pham 2025-05-24 11:04:14 -04:00
• b9018a3f9f [BugFix] Fix import error for fused_moe (#18642) wangxiyuan 2025-05-24 22:53:36 +08:00
• 4ceafb6299 [MISC] typo fix and clean import (#18664) Ning Xie 2025-05-24 22:52:09 +08:00
• 2e6705784f [CI/Build] chmod +x to cleanup_pr_body.sh (#18650) Cyrus Leung 2025-05-24 22:26:45 +08:00
• 1cb194a018 [Doc] Reorganize user guide (#18661) Cyrus Leung 2025-05-24 22:25:33 +08:00
• 2cd4d58df4 [Model] use AutoWeightsLoader for gpt2 (#18625) ztang2370 2025-05-24 21:36:13 +08:00
• 6d166a8d35 [Doc] Add community links (#18657) Cyrus Leung 2025-05-24 21:06:38 +08:00
• ef1dd6870f [Doc] Fix indentation problems in V0 Paged Attention docs (#18659) Cyrus Leung 2025-05-24 21:06:35 +08:00
• e77dc4bad8 [MISC][pre-commit] Add pre-commit check for triton import (#17716) Mengqing Cao 2025-05-24 20:09:15 +08:00
• 07458a51ce [Doc] Update README links, mark external links (#18635) Cyrus Leung 2025-05-24 17:57:15 +08:00
• c1e4a4052d [V1][Spec Decode] Support multi-layer eagle draft model (#18030) qizixi 2025-05-24 02:45:34 -07:00
• a859320575 [Model] Add support for Qwen2.5-Omni-7B-AWQ (Qwen2_5OmniForConditionalGeneration) (#18647) Yuanhao WU 2025-05-24 17:15:36 +08:00
• 441dc63ac7 [Frontend] improve vllm serve --help display (#18643) Reidandreidliu41 2025-05-24 15:53:22 +08:00
• d55e446d13 [V1][Spec Decode] Small refactors to improve eagle bookkeeping performance (#18424) qizixi 2025-05-23 23:51:22 -07:00
• ec82c3e388 FIX MOE issue in AutoRound format (#18586) Wenhua Cheng 2025-05-24 13:01:40 +08:00
• 45ab403a1f config.py: Clarify that only local GGUF checkpoints are supported. (#18623) Mathieu Borderé 2025-05-24 02:46:34 +02:00
• 2b10ba7491 [Bugfix][Nixl] Fix Preemption Bug (#18631) Robert Shaw 2025-05-23 19:30:16 -04:00
• 4fc1bf813a [Bugfix] Migrate to REGEX Library to prevent catastrophic backtracking (#18454) Feng XiaoLongandCrucifixion-Fxl 2025-05-24 07:16:26 +08:00
• f2036734fb [ModelOpt] Introduce VLLM_MAX_TOKENS_PER_EXPERT_FP4_MOE env var to control blockscale tensor allocation (#18160) Pavani Majety 2025-05-23 15:52:20 -07:00
• 7d9216495c [Doc] Update references to doc files (#18637) Cyrus Leung 2025-05-24 06:49:21 +08:00
• 0ddf88e16e [CI] Enable test_initialization to run on V1 (#16736) Michael Goin 2025-05-23 18:09:44 -04:00
• 1645b60196 Use prebuilt FlashInfer x86_64 PyTorch 2.7 CUDA 12.8 wheel for CI (#18537) Huy Do 2025-05-23 14:17:16 -07:00
• 2628a69e35 [V1] Support Deepseek MTP (#18435) Jiayi YaoandRui Qiao 2025-05-23 12:26:28 -05:00
• 371f7e4ca2 [Doc] Fix broken links and unlinked docs, add shortcuts to home sidebar (#18627) Cyrus Leung 2025-05-24 01:22:40 +08:00
• 15b45ffb9a [Doc] Avoid documenting dynamic / internal modules (#18626) Cyrus Leung 2025-05-24 00:58:02 +08:00
• 273cb3b4d9 [Doc] Fix top-level API links/docs (#18621) Cyrus Leung 2025-05-24 00:46:56 +08:00
• 8ddd1cf26a [Doc] fix list formatting (#18624) David Xia 2025-05-23 12:41:17 -04:00
• 6550114c9c [v1] Redo "Support multiple KV cache groups in GPU model runner (#17945)" (#18593) Chen Zhang 2025-05-24 00:39:47 +08:00
• 9520a989df [Docs] Change mkdocs to not use directory urls (#18622) Michael Goin 2025-05-23 12:33:21 -04:00
• 3d28ad343f Fix figures in design doc (#18612) Harry Mellor 2025-05-23 17:09:54 +01:00
• 6a7988c55b Refactor pplx init logic to make it modular (prepare for deepep) (#18200) youkaichao 2025-05-23 23:43:43 +08:00
• 022d8abe29 [Doc] Use a different color for the announcement (#18616) Cyrus Leung 2025-05-23 23:25:03 +08:00
• 5221815a00 [Doc] Fix markdown list indentation for MkDocs rendering (#18620) Hyogeun Oh (오효근) 2025-05-24 00:23:21 +09:00
• 1068556b2c [Bugfix][Build/CI] Fixup CUDA compiler version check for CUDA_SUPPORTED_ARCHS (#18579) Simon Mo 2025-05-23 07:43:58 -07:00
• 2cd1fa4556 [Misc] add Haystack integration (#18601) Reidandreidliu41 2025-05-23 21:21:19 +08:00
• d4c2919760 Include private attributes in API documentation (#18614) Harry Mellor 2025-05-23 15:18:31 +02:00
• 6220f3c6b0 [Bugfix] Fix transformers model impl ignored for mixtral quant (#18602) Tristan Leclercq 2025-05-23 14:54:13 +02:00
• 52fb23f47e Fix examples with code blocks in docs (#18609) Harry Mellor 2025-05-23 14:53:44 +02:00
• 6dd51c7ef1 [CI/Build] Fix V1 flag being set in entrypoints tests (#18598) Cyrus Leung 2025-05-23 20:51:53 +08:00
• 2edb533af2 Replace {func} with mkdocs style links (#18610) Harry Mellor 2025-05-23 14:51:38 +02:00
• 38a95cb4a8 [Doc] Fix indent of contributing to vllm (#18611) Hyogeun Oh (오효근) 2025-05-23 21:50:07 +09:00
• cd821ea5d2 [CI] fix kv_cache_type argument (#18594) Ning Xie 2025-05-23 19:49:18 +08:00
• 7ab056c273 [Hardware][CPU] Update intel_extension_for_pytorch 2.7.0 and move to requirements/cpu.txt (#18542) Kay Yan 2025-05-23 19:38:42 +08:00
• 6526e05111 Add myself as docs code owner (#18605) Harry Mellor 2025-05-23 13:08:31 +02:00
• e493e48524 [V0][Bugfix] Fix parallel sampling performance regression when guided decoding is enabled (#17731) Madeesh KannanandRussell Bryant 2025-05-23 12:38:23 +02:00
• 4ce64e2df4 [Bugfix][Model] Fix baichuan model loader for tp (#18597) Mengqing Cao 2025-05-23 17:39:05 +08:00
• fbb13a2c15 Revert "[V1] [Bugfix] eagle bugfix and enable correct lm_head for multimodal (#18034)" (#18600) Cyrus Leung 2025-05-23 17:18:22 +08:00
• a1fe24d961 Migrate docs from Sphinx to MkDocs (#18145) Harry Mellor 2025-05-23 11:09:53 +02:00
• d0bc2f810b [Bugfix] Add half type support in reshape_and_cache_cpu_impl on x86 cpu platform (#18430) Yuqi ZhangandYuqi Zhang 2025-05-23 01:41:37 -07:00
• b046cf792d [Feature][V1]: suupports cached_tokens in response usage (#18149) Chaunceyandsimon-mo 2025-05-23 16:41:03 +08:00
• 54af915949 [Doc] Update quickstart and install for cu128 using --torch-backend=auto (#18505) Michael Goin 2025-05-23 04:36:37 -04:00
• 71ea614d4a [Feature]Add async tensor parallelism using compilation pass (#17882) cascade 2025-05-23 01:03:34 -07:00
• 4c611348a7 [V1] [Bugfix] eagle bugfix and enable correct lm_head for multimodal (#18034) RonaldBXu 2025-05-23 00:37:18 -07:00
• 60cad94b86 [Hardware] correct method signatures for HPU,ROCm,XPU (#18551) Ning Xie 2025-05-23 13:31:59 +08:00
• 9c1baa5bc6 [Misc] Replace cuda hard code with current_platform (#16983) Shanshan Shen 2025-05-23 12:38:50 +08:00
• 4be2255c81 [Bugfix][Benchmarks] Fix a benchmark of deepspeed-mii backend to use api_key (#17291) Teruaki Ishizaki 2025-05-23 13:30:47 +09:00
• ed5d408255 [Neuron] Remove bypass on EAGLEConfig and add a test (#18514) aws-elaineyz 2025-05-22 21:26:32 -07:00
• 583507d130 [Spec Decode] Make EAGLE3 draft token ID mapping optional (#18488) Benjamin ChislettandWoosuk Kwon 2025-05-22 23:17:39 -04:00
• e44d8ce8c7 [Bugfix] Set KVTransferConfig.engine_id in post_init (#18576) lkchen 2025-05-22 19:54:42 -07:00
• 93ecb8139c [BugFix] Increase TP execute_model timeout (#18558) Nick Hill 2025-05-22 19:22:11 -07:00
• fae453f8ce [Misc] refactor: simplify input validation and num_requests handling in _convert_v1_inputs (#18482) CYJiang 2025-05-23 10:15:32 +08:00