Commit Graph

Select branches

Hide Pull Requests

main

patches

sm86

Mono Color

• d6249d0699 Fix typing for safetensors_load_strategy (#24641) Harry Mellor 2025-09-11 11:41:39 +01:00
• 25bb9e8c65 [CI Failure] fix models/language/pooling/test_auto_prefix_cache_support.py (#24636) wang.yuqi 2025-09-11 18:31:23 +08:00
• a1213fae5f [Misc] Add @NickLucche to codeowners (#24647) Nicolò Lucchesi 2025-09-11 11:18:09 +02:00
• a8b0361c92 [CI] Split pooling from entrypoints Test (#24632) wang.yuqi 2025-09-11 16:53:09 +08:00
• ed5ae4aace [Bugfix] Fix _synced_weight_loader (#24565) Kyuyeun Kim 2025-09-11 01:52:33 -07:00
• 0fc36463e0 [CI]Add transformers_utils to Async Engine, Inputs, Utils, Worker Test (#24615) Xingyu Liu 2025-09-11 01:52:10 -07:00
• d14c4ebf08 [Docs] Use 1-2-3 list for deploy steps in deployment/frameworks/ (#24633) Michael Yao 2025-09-11 16:50:12 +08:00
• ba6011027d [Docs] Update V1 doc to reflect whisper support (#24606) Russell Bryant 2025-09-11 04:50:08 -04:00
• 85df8afdae [Docs] Revise frameworks/anything-llm.md (#24489) Michael Yao 2025-09-11 16:50:05 +08:00
• 6aeb1dab4a [Bugfix] Fix incorrect import of CacheConfig (#24631) Cyrus Leung 2025-09-11 16:48:25 +08:00
• e93f4cc9e3 Add the support for the qwen3 next model (a hybrid attention model). (#24526) Tao HeandJee Jee Li 2025-09-11 15:32:09 +08:00
• 2048c4e379 [torchao] Support quantization configs using module swap (#21982) Jerry Zhang 2025-09-10 23:53:24 -07:00
• d13360183a Remove redundant all gather + split (#23441) 3 people 2025-09-10 23:45:07 -07:00
• 9bd831f501 [Model] New model support for Motif-1-Tiny (#23414) 3 people 2025-09-11 15:29:40 +09:00
• e2b1f863aa [Doc]: fixing doc typos (#24635) Didier Durand 2025-09-11 08:19:28 +02:00
• 41329a0ff9 [Core] feat: Add --safetensors-load-strategy flag for faster safetensors loading from Lustre (#24469) shengshiqi-googleandgemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> 2025-09-11 06:10:01 +00:00
• ee0bc5e1b4 Enable --profile in 'vllm bench throughput' (#24575) Tomas Ruiz 2025-09-11 08:06:19 +02:00
• 3d1393f6fc Kimi K2 Fused MoE kernels Optimization configs (#24597) Saman A. Pour 2025-09-10 23:06:16 -07:00
• 8a894084d2 [Engine][Chore] use local variable and remove output var assignment (#24554) Guy Stone 2025-09-11 02:05:42 -04:00
• e2d8c27f68 [BugFix] Fix pipeline parallel (#24621) Nick Hill 2025-09-10 23:05:30 -07:00
• 29799ddacc [Bugfix] Add missing VIT backend dispatch on CPU (#24623) Li, Jiang 2025-09-11 13:28:41 +08:00
• f17a6aa4ec [Ultravox] Fix Gemma instantiation, support quantization via --hf-overrides (#24131) Peter Salas 2025-09-10 22:25:34 -07:00
• 6c8deacd72 [Bug] [Spec Decode] Fix model_initialization test and mismatch in aux_hidden_layers (#24613) 3 people 2025-09-10 21:23:18 -07:00
• 55b823ba0f Add @chaunceyjiang to codeowner for reasoning Reasoning and Tool parser (#24406) Chauncey 2025-09-11 12:23:04 +08:00
• 8c5a747246 [distributed] update known issues (#24624) youkaichao 2025-09-11 11:09:38 +08:00
• 5931b7e5d9 [Models][Quantization] Add quantization configuration update in Voxtral model (#24122) 3 people 2025-09-10 22:13:56 -04:00
• cc99baf14d [Misc] Make timeout passable in init_distributed_environment (#24522) Jonathan Berkhahn 2025-09-10 15:41:12 -07:00
• dcb28a332b [Kernel] Flashinfer MLA (trtllm-gen) decode kernel integration (#21078) Hanjie QiuandMichael Goin 2025-09-10 18:31:10 -04:00
• fba7856581 [Perf] Warmup FlashInfer attention during startup (#23439) 4 people 2025-09-10 18:03:17 -04:00
• b5e383cd8b [gpt-oss] raise error for flashinfer backend without trtllm (#24482) Chen Zhang 2025-09-10 14:33:13 -07:00
• 9a161307f5 [torch.compile][ROCm][V1] Enable attention output FP8 fusion for V1 attention backends (#19767) 3 people 2025-09-10 16:59:55 -04:00
• 37e8182bfe [v1] Add Whisper model support (encoder-decoder) (#21088) Russell BryantandNickLucche 2025-09-10 16:53:35 -04:00
• 4db4426404 [CI] Fail subprocess tests with root-cause error (#23795) Nick Hill 2025-09-10 13:53:21 -07:00
• a0933c3bd6 [Bugfix] Enable FP8 KV cache for FlashInfer and Triton backend on non-sm100 GPUs (#24577) Thien Tran 2025-09-11 03:33:41 +08:00
• 09e68bce34 [Misc] update log level debug to warning when process port is used by (#24226) rongfu.lengandCyrus Leung 2025-09-11 02:32:57 +08:00
• 9fb74c27a7 [Core] Support configuration parsing plugin (#24277) Xingyu Liuandgemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> 2025-09-10 11:32:43 -07:00
• 4032949630 [Bugfix] Fix DeepEP config for DP4TP4 (#23619) Ming Yang 2025-09-10 10:37:56 -07:00
• 08abfa78ec [Bugfix] fix modelopt exclude_modules name mapping (#24178) tomeras91andCyrus Leung 2025-09-10 20:20:46 +03:00
• 2bef2d1405 [Logging] allow config logging stream (#24336) Shiyan Deng 2025-09-10 08:02:01 -07:00
• 36cacd0958 [Doc] Add documentation for GLM-4.5 series models: tool-calling and reasoning parser (#24589) Robin 2025-09-10 22:50:55 +08:00
• bb3eb80d92 [Core] Split LoRA layers (#24574) Jee Jee Li 2025-09-10 22:47:51 +08:00
• fcc0a3130a [CI] Fix tensorizer test assertion (#24545) pwschuurman 2025-09-10 06:57:36 -07:00
• 736569da8d [Platform] Custom ops support for LMhead and LogitsProcessor (#23564) zzhxxx 2025-09-10 21:26:31 +08:00
• 2eb9986a2d [BugFix] python collect_env.py and vllm collect-env compatibility with uv venv (#24066) Kay Yan 2025-09-10 21:25:33 +08:00
• ccee371e86 [Docs] Fix warnings in mkdocs build (continued) (#24092) 3 people 2025-09-10 22:23:28 +09:00
• c0bd6a684a Fix Auto_Round Quatization Loading on SM75 and Lower GPUs (#24217) RoadToNowhereXandWentao Ye 2025-09-10 23:22:31 +10:00
• 3144d90217 fix some typos (#24167) co63ocandRussell Bryant 2025-09-10 21:21:23 +08:00
• 2f5e5c18de [CI/Build] bump timm dependency (#24189) DanieleandRussell Bryant 2025-09-10 15:20:59 +02:00
• bd98842c8a [CI] Add PPL test for generation models (#24485) wang.yuqi 2025-09-10 21:16:39 +08:00
• d6069887c6 [rocm] enable torchao quantization for rocm (#24400) Lifans 2025-09-10 06:16:21 -07:00
• 492196ed0e [CI/Build] split true unit tests to Entrypoints Unit Tests (#24418) Ye (Charlotte) Qi 2025-09-10 06:16:07 -07:00
• f4f1a8df22 [BugFix] Ensure integrity of reused CPU tensors during async scheduling (#24527) Nick Hillandguoze.lin 2025-09-10 06:15:14 -07:00
• 0b9a612fa3 [BugFix][easy] Fix flaky test test_gpt_oss_multi_turn_chat (#24549) lacoraandlacora2017 2025-09-10 06:14:55 -07:00
• 4c04eef706 [BugFix][Multi Modal] Fix TensorSchema shape mismatch in Molmo (#24559) Wenlong Wang 2025-09-10 06:14:27 -07:00
• f36355abfd Move LoadConfig from config/__init__.py to config/load.py (#24566) Harry Mellor 2025-09-10 14:14:18 +01:00
• 9e3c3a7df2 [LoRA]: Add LoRA support to Mistral's Voxtral models (#24517) Yash Pratap SinghandJee Jee Li 2025-09-10 18:42:03 +05:30
• 6cbd41909e Feature/vit attention unification# 23880 (#23978) baonudesifeizhaiandIsotr0py 2025-09-10 09:10:14 -04:00
• 72d30108a0 Support for NemotronH Nano VLM (#23644) danielafrimi 2025-09-10 16:10:06 +03:00
• 8b83b93739 [Docs] Document the extra memory footprint overhead when using EPLB (#24537) Tyler Michael Smith 2025-09-10 09:09:49 -04:00
• 9dbefd88e9 [Docs] Improve organisation of API Reference nav (#24569) Harry Mellor 2025-09-10 14:08:21 +01:00
• 7c195d43da [ROCm][Bugfix] Fix Aiter RMSNorm (#23412) vllmellm 2025-09-10 21:08:03 +08:00
• 0ae43dbf8c [Attention] add DCP support for FLASH_ATTN_MLA backend (#24453) Lucas WilkinsonandMatthew Bonanni 2025-09-10 05:19:26 -04:00
• 267c80d31f [Model] Limit CPU threads for image transformations in InternVL to reduce cpu contention. (#24519) li-jinpengandRoger Wang 2025-09-10 16:45:44 +08:00
• 77f62613f9 Consolidate rendering parameters into RenderConfig dataclass (#24543) Flora Feng 2025-09-10 01:44:47 -07:00
• feaf202e93 [Bugfix] Guard _may_reorder_batch for encoder-only models on CPU (#24319) (#24348) RemyandLi, Jiang <jiang1.li@intel.com> 2025-09-10 15:24:42 +09:00
• 91130ae376 [docs] promo pytorch conf and ray summit (#24562) Simon Mo 2025-09-09 23:24:20 -07:00
• e40827280b [Docs] Enable relative links in examples to function when rendered in the docs (#24041) Harry Mellor 2025-09-10 05:40:45 +01:00
• 4377b1ae3b [Bugfix] Update Run:AI Model Streamer Loading Integration (#23845) 3 people 2025-09-09 21:37:17 -07:00
• 009d689b0c [Core] Simplify and unify mm uuid handling & auto-generated mm hash overrides processing. (#24271) Chenheli Hua 2025-09-09 21:36:09 -07:00
• 0efdb5c3ba [gpt-oss] Cache permute indices for faster MXFP4 MoE layer loading (#24154) Wei 2025-09-09 21:27:53 -07:00
• 53b42f4102 [BugFix][Spec Decode] Fix out-of-range index triggered by eagle3; re-enable test for LlamaForCausalLMEagle3 (#24392) Wenlong Wang 2025-09-09 21:24:23 -07:00
• 309d7aa401 [P/D] MultiConnector supports shutdown (#24425) Chauncey 2025-09-10 12:24:11 +08:00
• b4a01aaf95 [KV Connector] More async support for get_num_new_matched_tokens (#23620) Yihua Cheng 2025-09-09 21:23:37 -07:00
• 83dd28aae4 [CI] Adjust threshold for flaky ngram spec decoding test (#24528) Nick Hill 2025-09-09 21:07:33 -07:00
• f88e84016f [BugFix] Fix async core engine client finalizer (#24540) Nick Hill 2025-09-09 21:07:13 -07:00
• 3c2156b3af [Hardware][Apple-CPU] Enable native bfloat16 on Apple Silicon (M2 and later) (#24129) Ignacio Sica 2025-09-10 00:50:21 -03:00
• 7e7db04310 [CI] Retry flaky fp8 cutlass mla tests (#24536) Nick Hill 2025-09-09 20:33:10 -07:00
• 41f160b974 Add @heheda12345 to CODEOWNERS of KVCacheManager related code (#24546) Chen Zhang 2025-09-09 20:30:32 -07:00
• dc625ea6b8 [Perf] Convert np array to torch tensor to index into block table for attn chunking (#24474) Yong Hoon Shin 2025-09-09 20:01:06 -07:00
• b23fb78623 [Bugfix] Fix for 24530. Fix naive all2all shared expert overlap. (#24538) bnellnm 2025-09-09 20:53:53 -04:00
• 561f38dc3c [Bugfix] Improve EPLB config validation error message (#24524) Tyler Michael Smith 2025-09-09 20:32:36 -04:00
• 73e688cb79 [ROCm][Feature] Enable Pipeline Parallelism with Ray Compiled Graph on ROCm (#24275) Charlie Fu 2025-09-09 18:27:35 -05:00
• fb1a8f932a [Benchmark] Add option to skip oversampling in benchmark (#24457) Ekagra Ranjan 2025-09-09 18:00:17 -04:00
• 0dc9cbb527 [Benchmark] Update bench doc with mtbench, blazedit, spec bench (#24450) Ekagra Ranjan 2025-09-09 17:15:41 -04:00
• b5fb3005a8 [Log] Use a relative path in debug-level logs to distinguish files with identical names (#23846) Jiangyun Zhu 2025-09-10 04:46:35 +08:00
• 15de5ff9ea [Feature] Disallow FlashMLA on Blackwell (#24521) Wentao Yeandgemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> 2025-09-09 14:59:34 -04:00
• b8a93076d3 [CI] execute all piecewise compilation tests together (#24502) Jiangyun Zhu 2025-09-10 02:05:25 +08:00
• c3f9773b2c [TPU] Fix tpu structured decoding in mixed batches (#24458) Chenyaaang 2025-09-09 23:34:25 +05:30
• 3707cb2505 [Docs] Gemma3n transcriptions endpoint support (#24512) Nicolò Lucchesi 2025-09-09 20:03:32 +02:00
• 920ed46b09 [Misc] bump outlines_core to fix the version conflicts with outlines >= 1.2.0 (#24368) 3 people 2025-09-10 02:59:46 +09:00
• 15cb047e25 Extend renderer with embedding support and integrate completion endpoint (#24405) Flora Feng 2025-09-09 10:46:46 -07:00
• 9ad0688e43 [Bugfix] Fix hidden_size for multimodal classification model (#24501) Jee Jee Li 2025-09-10 01:37:25 +08:00
• b9a1c4c8a2 [ROCm][CI/Build] Sync ROCm dockerfiles with the ROCm fork (#24279) Gregory Shtrasberg 2025-09-09 12:21:56 -04:00
• 1aa427fdc1 [Kernels] Add Flash Linear Attention Kernels (#24518) youkaichao 2025-09-10 00:04:41 +08:00
• 1c63a16b65 [Core] Run garbage collector after CUDA graph capture to fix throughput regression (#24128) Micah WilliamsonandGregory Shtrasberg 2025-09-09 09:38:10 -05:00
• 922d3b401b [Bugfix] Handle the edge case in detokenizer where processed tokens contain both stop str and eos token (#23938) d.transposed 2025-09-09 16:30:24 +02:00
• 19332c0479 [Model] Systematic support for fp32 head, pooling models part (#23810) wang.yuqi 2025-09-09 22:29:50 +08:00
• a55cf41a09 [Compilation][WideEP] Enable Piecewise CUDAGraph for DeepEPHT (#24123) Wentao Ye 2025-09-09 10:21:10 -04:00
• 6fb2788163 [CI/Build][Doc] Fully deprecate old bench scripts for serving / throughput / latency (#24411) Ye (Charlotte) Qi 2025-09-09 03:02:35 -07:00
• 3d2a2de8f7 [RL] fast weight update with zmq + ipc handles (#24295) Weixiao Huangandyoukaichao 2025-09-09 16:57:46 +08:00