Commit Graph

Select branches

Hide Pull Requests

main

patches

sm86

Mono Color

• e471d7ca7e [CI/Build][Bugfix] fix qutlass cmake error when set QUTLASS_SRC_DIR (#26773) zhrrrandMichael Goin 2025-10-15 12:09:44 +08:00
• c43ca8259e [Docs] Move build.inc into arm.inc (#26862) Michael Yao 2025-10-15 11:35:08 +08:00
• 85a65e7f51 [Model] Add DeepSeek-V3.1 reasoning parser (split from PR #24972) (#25589) 3 people 2025-10-15 11:09:52 +08:00
• a2986b3e33 [Bugfix] Fixes prefix-repetition benchmark script (#26828) kourosh hakhamaneshi 2025-10-14 19:54:43 -07:00
• 96b9aa5aa0 [Frontend][torch.compile] CompilationConfig Overhaul (#20283): name change compilation level to compilation mode, deprecation compilation level (#26355) Morrison TurnanskyandLuka Govedič 2025-10-14 22:51:16 -04:00
• e66d787bce Disable FlashInfer sampler by default (#26859) Michael Goin 2025-10-14 22:35:18 -04:00
• bfad142e25 [BUGFIX][NIXL] quick fix for 'assert self.connector_worker is not None' in get_kv_connector_stats (#26851) Chendi.Xue 2025-10-14 21:33:25 -05:00
• 9354660036 [Bugfix]fix Qwen3 xml tool parser (#26345) Zhikaiiii 2025-10-15 09:50:30 +08:00
• 07ca70af8d [Core][Easy] Use envs.__getattr__ for all Unify to environment variable access (#26810) Jialin Ouyang 2025-10-14 18:41:18 -07:00
• 2dcd12d357 [torch.compile] Fix tests for torch==2.9 inductor partition (#26116) Luka Govedič 2025-10-14 19:55:02 -04:00
• 579d2e5458 [WideEP][P/D] Add usage stats for DP+EP and KV Connector (#26836) Tyler Michael Smith 2025-10-14 19:51:54 -04:00
• 0512c04aee [frontend][gptoss] Add per turn stats into Harmony Context (#25061) Ye HuandYe Hu 2025-10-14 16:48:13 -07:00
• 7e0ef4084a [CI Failure] Fix torchao dep failure for Quantization Test (#26824) Michael Goin 2025-10-14 19:41:43 -04:00
• 4aed506b65 [Core] Streamline some structured output related code (#26737) Nick Hill 2025-10-14 16:27:44 -07:00
• a86b4c58e8 remove attn output view kernel (#26680) Boyuan Fengandgemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> 2025-10-14 15:53:10 -07:00
• ff4810ba73 [Minor] Group async_scheduling related fields in model runner init (#26736) Nick Hill 2025-10-14 14:46:37 -07:00
• 9d6964926e fix: response_format for completion (#23212) Nan Qin 2025-10-14 16:23:22 -05:00
• 0e65818910 Added MoE configs for llama 4, H200 device with tp=4/8 tuning (#26837) Dhruvil Bhatt 2025-10-14 14:21:03 -07:00
• 380f17527c [Perf] Cache vllm.env.__getattr__ result to avoid recomputation (#26146) Jialin Ouyang 2025-10-14 14:03:21 -07:00
• b92ab3deda Notice for deprecation of AutoAWQ (#26820) HDCharlesandgemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> 2025-10-14 16:39:59 -04:00
• acaa2c0a4a [Core] Reuse empty block lists whenever possible in KVCacheBlocks to mitigate GC costs (#24964) Jialin Ouyang 2025-10-14 12:58:43 -07:00
• 82af928c41 [Attention][Spec Decode] FlashMLA spec decode support (#26541) Matthew Bonanni 2025-10-14 15:38:20 -04:00
• 87efc681db llama4_vision_rope: add HIP override to accept (q, k) and avoid (positions, q, k) mismatch (#26790) Huamin Li 2025-10-14 11:54:12 -07:00
• c3a722fcb2 [CI Failure] Fix tests with missing TinyLlama-1.1B-Chat-v1.0-FP8-e2e (#26816) Michael Goin 2025-10-14 14:38:59 -04:00
• aba48f7db1 [Kernel][MoE] Add MoE tunings for GLM 4.6-FP8 and GLM 4.5 Air on NVidia B200 (#26818) Ze'ev Klapow 2025-10-14 14:20:39 -04:00
• 04b5f9802d [CI] Raise VLLM_MAX_SIZE_MB to 500 due to failing Build wheel - CUDA 12.9 (#26722) Michael Goin 2025-10-14 13:52:05 -04:00
• efc8f7d814 Update coveragerc and add codecov.yml for path fixes (#26435) Reza Barazesh 2025-10-14 12:45:06 -04:00
• 6d87a2838c [Config] Remove Unused Environment Variable VLLM_DISABLE_PAD_FOR_CUDAGRAPH (#26743) Wentao Ye 2025-10-14 11:47:49 -04:00
• e6cdbd6792 Revert "[issues template] Encourage the author implement their own ideas" (#26814) wang.yuqi 2025-10-14 23:37:34 +08:00
• df850c4912 [Feature][Responses API] Stream Function Call - harmony (#24317) Chauncey 2025-10-14 23:31:43 +08:00
• 720394de43 [KVConnector][Metrics] Aggregate scheduler-side KVConnectorStats (#26046) Qier Li 2025-10-14 10:38:07 -04:00
• 88a49745af [issues template] Encourage the author implement their own ideas (#26671) wang.yuqi 2025-10-14 22:32:36 +08:00
• ca683a2a72 use combo kernel to fuse qk-norm and qk-rope (#26682) Boyuan Feng 2025-10-14 06:40:59 -07:00
• e9f1b8c9e9 Adjusted the model order of the model registration file (#26798) 汪志鹏 2025-10-14 21:26:11 +08:00
• ea97940d6c [DCP] Support Decode Context Parallel (DCP) for GQA with FlashAttention (#24864) Jaya Yuan 2025-10-14 21:07:50 +08:00
• fdd32750f0 [CI/Build] Cleanup LoRA test (#26752) Jee Jee Li 2025-10-14 20:06:35 +08:00
• c715ba3735 [Feature] Change vllm.py with pydantic validation (#26726) Vladislav BronzovandHarry Mellor 2025-10-14 14:00:54 +02:00
• 9c4cb68339 [Chore] Remove SupportsV0Only interface and update supported models docs (#26783) Cyrus Leung 2025-10-14 19:55:10 +08:00
• 780eb03d9b [CI] Fix test_tool_id_kimi_k2 (#26787) Chauncey 2025-10-14 18:27:07 +08:00
• ef9676a1f1 [Doc] ruff format some Python examples (#26767) Cyrus Leung 2025-10-14 18:21:53 +08:00
• 70b1b330e1 Don't allow typos to fix by default (#26785) Harry Mellor 2025-10-14 11:05:15 +01:00
• d1d063a588 [Chore] Use max_transformers_version for Qwen-VL test (#26792) Cyrus Leung 2025-10-14 18:03:46 +08:00
• 7e6edb1469 [NIXL][HeteroTP] Enable KV transfer from HND prefill to NHD decode (#26556) Chendi.Xue 2025-10-14 04:46:05 -05:00
• 74704d4553 [Model] Use merge_by_field_config for MM models (O-P) (#26776) Cyrus Leung 2025-10-14 17:42:45 +08:00
• d2f816d6ff [Bugfix] Standardize merging multimodal embeddings (#26771) Cyrus Leung 2025-10-14 17:36:21 +08:00
• 577d498212 [Plugin] Make plugin group clear (#26757) wangxiyuan 2025-10-14 15:49:59 +08:00
• fd85c9f426 [Bugfix][FE]: Always include usage with --enable-force-include-usage (#20983) Max WittigandAntoine Auger 2025-10-14 09:17:39 +02:00
• d32c611f45 [CI/Build] Use 127.0.0.1 instead of localhost in utils (#26750) Ye (Charlotte) Qi 2025-10-14 00:04:00 -07:00
• 01ad27faff [Model][Bugfix]fix ernie45 load failed due to ernie45 eplb code (#26684) CSWYF3634076 2025-10-14 14:55:23 +08:00
• 481545b397 scheduler.py: Update the name of the default scheduler. (#26758) Ryan Li 2025-10-14 14:52:21 +08:00
• d3cc8427c0 [ci] Adding the test-amd.yaml for test definitions for the AMD backend. (alternative PR) (#26718) Alexei-V-Ivanov-AMD 2025-10-14 01:10:23 -05:00
• 4821ac1b4d [CI] [ROCm] Automate CC list for ROCm related issue (#26753) vllmellm 2025-10-14 13:57:26 +08:00
• 4497c8f821 Fix lora tests failure in TPU CI due to the removal of LoRA bias (#26723) XiongfeiWei 2025-10-13 22:04:23 -07:00
• 2e36cdbe2b [Docs] Add a start tag to build.inc.md (#26747) Michael Yao 2025-10-14 12:51:55 +08:00
• fe3edb4cf0 Add support for the /rerank endpoint in vllm bench serve (#26602) Maximilien de Bayser 2025-10-14 01:25:43 -03:00
• 29350922c6 [Feature][Quantization] auto_round format add support for regex (#24024) Heng Guoandgemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> 2025-10-14 11:03:16 +08:00
• 8ae169286f [torch.compile] Unwrap fused_marlin_moe custom op (#26739) Varun Sundar RabindranathandVarun Sundar Rabindranath 2025-10-13 22:22:16 -04:00
• 8a0af6a561 [build][torch.compile] upgrade depyf version (#26702) youkaichao 2025-10-14 10:12:09 +08:00
• cfded80793 [Easy] Fix env type check errors from VLLM_DEBUG_LOG_API_SERVER_RESPONSE (#26742) Jialin Ouyang 2025-10-13 18:46:44 -07:00
• b59dd19b55 [compile] Enable sequence parallelism for full cuda graph without specifying compile sizes (#26681) Angela Yi 2025-10-13 18:15:34 -07:00
• 3e051bda82 [UX] Replace VLLM_ALL2ALL_BACKEND with --all2all-backend (#26732) Michael Goin 2025-10-13 21:12:52 -04:00
• 8317f72354 [Misc][DP] support customized aggregated logger for dp (#24354) Lucia Fang 2025-10-13 17:45:59 -07:00
• d8bebb008a Add tests for chunked prefill and prefix cache with causal pooling models (#26526) Maximilien de BayserandAyush Singh 2025-10-13 20:45:04 -03:00
• 35bc22f23c [ResponseAPI] Further polish message serialization and unit tests (#26728) Jialin Ouyang 2025-10-13 16:31:35 -07:00
• fa96fb9c70 Pruning kernel Core Tests (#26727) Fardin Hoque 2025-10-13 16:08:18 -07:00
• e3fdb627d9 [FrontEnd] UNREVERT CompilationConfig overhaul (#20283): deprecate use_inductor in favor of backend, simplify custom_ops (#26502) 3 people 2025-10-13 18:47:16 -04:00
• 7200a21cd1 [Bug] Fix Assertion error DeepEP/csrc/kernels/intranode.cu:928: 'false and Unsupported type' (#26532) Wentao Ye 2025-10-13 18:26:37 -04:00
• 577c72a227 [CI Perf]Prune Tests in kernel/mamba (#26538) Fardin HoqueandWentao Ye 2025-10-13 15:22:31 -07:00
• 314285d4f2 [CI] Fix mypy for vllm/distributed (#26593) Wentao YeandHarry Mellor 2025-10-13 16:02:24 -04:00
• d2a7938582 [Frontend][1/N] Improve all pooling task | Support FP16 Embedding Base64 (Still uses fp32 by default). (#26414) 3 people 2025-10-14 03:06:43 +08:00
• 89342ce4c0 [Quantization] [Performance] Enable Marlin GEMM kernels for the calibration-free RTN-based quantization (#26051) Alex Kogan 2025-10-13 14:52:54 -04:00
• f89f599395 [CI][Release][Arm64]: Build arm64 release for gpu arch 8.9 (#26698) Yibo Cai 2025-10-14 02:42:12 +08:00
• e251e457c5 [Log] Optimize Startup Log (#26601) Wentao Ye 2025-10-13 14:06:57 -04:00
• afc47e4de7 [Model] Use merge_by_field_config for MM models (M-N) (#26710) Cyrus Leung 2025-10-14 01:27:01 +08:00
• e3b90c1ba2 [Bugfix][Speculative Decoding] Extend Eagle quantization config fix to llama_eagle.py (#26590) Rahul Tuli 2025-10-13 22:47:13 +05:30
• 134f70b3ed [Bugfix][Rocm] fix qr error when different inp shape (#25892) 3 people 2025-10-14 01:04:21 +08:00
• a1b2d658ee [CI/Build] upgrade compressed-tensors to 0.12.2 to address LGPLv3 (#26501) Sangyeon Cho 2025-10-14 01:58:33 +09:00
• 5c7fe25491 [Misc] Separate prompt logging to debug (#26713) Aleksei Tsvetkov 2025-10-13 19:04:18 +03:00
• 53c9a7cee2 [P/D] [NixlConnector] kv load recovery integration (#26171) Will Eaton 2025-10-13 11:48:04 -04:00
• 0d21b9b51e [UX] Speedup DeepGEMM warmup with heuristics (#25619) 3 people 2025-10-13 10:59:27 -04:00
• 10214b6935 [FEATURE]: Use pydantic validation in multimodal.py config (#26629) Anand RoyandHarry Mellor 2025-10-13 20:26:59 +05:30
• 4a61950f4d [Hardware][CPU] Disable torch.compile for RISC-V to prevent APIError (#26693) ihb2032 2025-10-13 22:56:01 +08:00
• 3263799056 [unrevert] Add batch invariant kernel override for FlashInfer backend [2/n] (#26373) Bram Wasti 2025-10-13 07:24:53 -07:00
• 8e67b2557a [Bugfix] Fix out of bound index issue for Jina-embedding-v3 RoPE with cuda graph (#26687) Isotr0py 2025-10-13 18:21:48 +08:00
• 4073c82c4e [ResponseAPI] Simplify input/output message serialization (#26620) Jialin Ouyang 2025-10-13 02:59:15 -07:00
• 767c3ab869 [Model][0/N] Improve all pooling task | clean up (#25817) wang.yuqi 2025-10-13 16:44:50 +08:00
• 4f207c7174 Ignore large reformatting PRs in git blame (#26690) Harry Mellor 2025-10-13 09:20:47 +01:00
• 782505ed8e [Model] Add reasoning_parser and tool_parser for Ernie45 thinking (#25027) CSWYF3634076 2025-10-13 15:55:20 +08:00
• 98f30b8cba [Model] Fix Skywork R1V mlp (#26673) Jee Jee Li 2025-10-13 13:42:17 +08:00
• 3cd36660f7 docs: wrong command in structured_outputs README (#26677) yihong 2025-10-13 11:59:01 +08:00
• 46ad73955a [FIX] Throwing an exception when the model does not support pool tasks (#25840) (#25855) yyzxwandwang.yuqi 2025-10-13 11:56:21 +08:00
• 41f3884438 [Bugfix][Core]Fix block table out-of-range issue in priority scheduling (#26661) quanliu 2025-10-13 09:25:42 +08:00
• 60e419c1ee [Misc] cache result of disable_inplace (#26666) bnellnm 2025-10-12 20:17:50 -04:00
• 7ef6052804 [CI/Build] Add tool to build vllm-tpu wheel (#19165) Michael Goin 2025-10-12 18:25:40 -04:00
• 4fca1a1bd2 [easy] fix pre commit error on trunk (#26665) Huamin Li 2025-10-12 14:25:34 -07:00
• a6049be73c [Models][Qwen3VL] Speedup fast_pos_embed_interpolate (#26647) Lukas Geiger 2025-10-12 18:20:07 +01:00
• 18ed7746ea [Feature] Add support for naver/splade-v3 (BERT-based sparse embedding model) (#26339) gjgjosandCyrus Leung 2025-10-13 02:00:52 +09:00
• 8fcaaf6a16 Update Optional[x] -> x | None and Union[x, y] to x | y (#26633) Harry Mellor 2025-10-12 17:51:31 +01:00
• 9bb38130cb [Bugfix] Fix GPU_ID issue in test script (#26442) Chendi.Xue 2025-10-12 06:39:05 -05:00
• b91d8db873 [Bugfix][DCP] Set default CUDAGraphMode to PIECEWISE for DCP (#26574) Jaya Yuan 2025-10-12 17:58:38 +08:00