Commit Graph

Select branches

Hide Pull Requests

main

patches

sm86

Mono Color

• 4c33d67321 [Bugfix] fix tmp_out and exp_sums dimensions (#17438) Hui Liu 2025-05-02 09:44:07 -07:00
• cb234955df [Misc] Clean up input processing (#17582) Cyrus Leung 2025-05-02 23:11:53 +08:00
• 3a500cd0b6 [doc] miss result (#17589) Reidandreidliu41 2025-05-02 22:04:49 +08:00
• 868c546da4 Support W8A8 INT8 MoE for compressed-tensors (#16745) Michael Goin 2025-05-02 08:03:32 -06:00
• 99404f53c7 [Security] Fix image hash collision (#17378) Cyrus Leung 2025-05-02 20:36:39 +08:00
• 785d75a03b Automatically tell users that dict args must be valid JSON in CLI (#17577) Harry Mellor 2025-05-02 13:24:55 +01:00
• 6d1479ca4b [doc] add the print result (#17584) Reidandreidliu41 2025-05-02 20:24:45 +08:00
• b8b0859b5c add more pytorch related tests for torch nightly (#17422) Yang Wang 2025-05-02 03:29:59 -07:00
• d7543862bd [Misc] Rename assets for testing (#17575) Cyrus Leung 2025-05-02 18:29:25 +08:00
• c777df79f7 [BugFix] Fix Memory Leak (#17567) Robert Shaw 2025-05-02 04:07:03 -04:00
• cc2a77d7f1 [Core] [Bugfix] Add Input Embeddings (#15428) 6 people 2025-05-02 03:06:39 -05:00
• 9e2de9b9e9 [Bugifx] Remove TritonPlaceholder from sys.modules (#17317) Isotr0py 2025-05-02 15:45:01 +08:00
• 109e15a335 Add pt_load_map_location to allow loading to cuda (#16869) Jerry Zhang 2025-05-01 23:23:42 -07:00
• f192ca90e6 Fix PixtralHF missing spatial_merge_size (#17571) Michael Goin 2025-05-01 23:14:09 -06:00
• f89d0e11bf [Misc] Continue refactoring model tests (#17573) Cyrus Leung 2025-05-02 13:06:08 +08:00
• b4003d11fc Check if bitblas is installed during support check (#17572) Michael Goin 2025-05-01 22:32:54 -06:00
• 292fc59d61 [CI] Actually run tests/kv_transfer/test_disagg.py in CI (#17555) Michael Goin 2025-05-01 22:05:04 -06:00
• afcb3f8863 [Attention] MLA move o_proj q_proj into cuda-graph region (#17484) Lucas Wilkinson 2025-05-01 23:16:26 -04:00
• afb12e4294 [Doc] note that not all unit tests pass on CPU platforms (#17554) David Xia 2025-05-01 22:57:21 -04:00
• 24aebae177 [Bugfix] Disable gptq_bitblas for <SM80 to fix GPTQ on V100/T4 (#17541) Michael Goin 2025-05-01 18:59:35 -06:00
• 39c0813a7f [V1][Spec Decode] Apply torch.compile & cudagraph to EAGLE3 (#17504) qizixi 2025-05-01 16:19:30 -07:00
• 9b70e2b4c1 [Misc][Tools][Benchmark] Publish script to auto tune server parameters (#17207) Chenyaaang 2025-05-01 12:53:03 -07:00
• 173daac19d [Bug]change the position of cuda_graph_sizes in dataclasses (#17548) Chen Xia 2025-05-01 11:52:37 -07:00
• 04f2cfc894 Remove duplicate code from dbrx.py (#17550) sstamenk 2025-05-01 20:51:58 +02:00
• 811a6c0972 [ROCM] Add gfx950 to the custom attention archs (#16034) 3 people 2025-05-01 13:18:28 -05:00
• 9b1769dd9a [Bugfix] Fix lint error (#17547) Cyrus Leung 2025-05-02 02:12:19 +08:00
• 61c299f81f [Misc]add configurable cuda graph size (#17201) 3 people 2025-05-01 11:04:50 -07:00
• 4acfa3354a [ROCm] update installation guide to include build aiter from source instructions (#17542) Hongxia YangandCyrus Leung 2025-05-01 14:01:28 -04:00
• 88c8304104 [Model] Refactor Ovis2 to support original tokenizer (#17537) Isotr0py 2025-05-02 02:00:53 +08:00
• 6768ff4a22 Move the last arguments in arg_utils.py to be in their final groups (#17531) Harry Mellor 2025-05-01 18:31:44 +01:00
• f2e7af9b86 [CI/Build] Remove awscli dependency (#17532) Cyrus Leung 2025-05-02 00:20:54 +08:00
• 7423cf0a9b [Misc] refactor example - cpu_offload_lmcache (#17460) Reidandreidliu41 2025-05-01 23:05:24 +08:00
• 460a2b1100 [torch.compile] Add torch inductor pass for fusing silu_and_mul with subsequent scaled_fp8_quant operations (#10867) Sage Moore 2025-05-01 07:59:28 -07:00
• 28566d73b3 [ROCm] remove unsupported archs from rocm triton flash-attention supported list (#17536) Hongxia Yang 2025-05-01 10:54:25 -04:00
• 98060b001d [Feature][Frontend]: Deprecate --enable-reasoning (#17452) Chauncey 2025-05-01 21:46:16 +08:00
• f5a3c655b2 [FEAT] [ROCm]: Add Qwen/Qwen3-235B-A22B-FP8 TP4 triton fused moe config (#17535) TJian 2025-05-01 21:37:17 +08:00
• 7169f87ad0 [doc] add streamlit integration (#17522) Reidandreidliu41 2025-05-01 21:34:02 +08:00
• b74d888c63 Fix more broken speculative decode tests (#17450) Huy Do 2025-05-01 06:05:58 -07:00
• 2007d4d54f [FEAT] [ROCm]: Add Qwen/Qwen3-30B-A3B-FP8 fused moe config for MI300X (#17530) TJian 2025-05-01 21:03:13 +08:00
• 48e925fab5 [Misc] Clean up test docstrings and names (#17521) Cyrus Leung 2025-05-01 20:19:32 +08:00
• 1903c0b8a3 [Frontend] Show progress bar for adding requests (#17525) Cyrus Leung 2025-05-01 20:15:32 +08:00
• 86a1f67a3b [Bugfix][Benchmarks] Allow benchmark of deepspeed-mii backend to select a model (#17285) Teruaki Ishizaki 2025-05-01 20:54:51 +09:00
• a257d9bccc Improve configs - ObservabilityConfig (#17453) Harry Mellor 2025-05-01 11:52:05 +01:00
• 015069b017 [Misc] Optimize the Qwen3_ReasoningParser extract_reasoning_content (#17515) Chauncey 2025-05-01 18:29:01 +08:00
• fbefc8a78d [Core] Enable IPv6 with vllm.utils.make_zmq_socket() (#16506) Russell Bryant 2025-05-01 05:38:18 -04:00
• 26bc4bbcd8 Avoid overwriting vllm_compile_cache.py (#17418) Keyun Tong 2025-05-01 00:30:57 -07:00
• 3c3d767201 [BugFix] Fix mla cpu - missing 3 required positional arguments (#17494) Lucas Wilkinson 2025-05-01 02:36:52 -04:00
• 13cf6b6236 [BugFix] fix speculative decoding memory leak when speculation is disabled (#15506) Noah Yoshida 2025-04-30 23:28:17 -07:00
• 90d0a54c4d [ROCm] Effort to reduce the number of environment variables in command line (#17229) Hongxia Yang 2025-05-01 02:27:06 -04:00
• 7a0a146c54 [Build] Require setuptools >= 77.0.3 for PEP 639 (#17389) Russell Bryant 2025-05-01 02:25:36 -04:00
• 7ab643e425 FIxing the AMD test failures caused by PR#16457 (#17511) Alexei-V-Ivanov-AMD 2025-05-01 01:23:07 -05:00
• afb4429b4f [CI/Build] Reorganize models tests (#17459) Cyrus Leung 2025-05-01 14:03:08 +08:00
• aa4502e7f3 [CI][Bugfix] Fix failing V1 Test due to missing 'cache_salt' arg (#17500) Michael Goin 2025-04-30 22:03:30 -06:00
• 17b4d85f63 [CI][TPU] Skip structured outputs+spec decode tests on TPU (#17510) Michael Goin 2025-04-30 21:36:20 -06:00
• 1144a8efe7 [Bugfix] Temporarily disable gptq_bitblas on ROCm (#17411) NaLan ZeYu 2025-05-01 10:51:45 +08:00
• 08fb5587b4 [Bugfix][ROCm] Fix import error on ROCm (#17495) Gregory Shtrasberg 2025-04-30 22:51:42 -04:00
• dbc18e7816 [CI][TPU] Skip Multimodal test (#17488) Siyuan Liu 2025-04-30 19:51:39 -07:00
• 02bd654846 [Misc] Rename Audios -> Audio in Qwen2audio Processing (#17507) Alex Brooks 2025-04-30 20:51:36 -06:00
• 200bbf92e8 Bump Compressed Tensors version to 0.9.4 (#17478) Rahul Tuliandmgoin 2025-04-30 17:24:45 -05:00
• 81ecf425f0 [v1][Spec Decode] Make sliding window compatible with eagle prefix caching (#17398) Chen Zhang 2025-05-01 02:25:53 +08:00
• 42d9a2c4c7 doc: fix bug report Github template formatting (#17486) David Xia 2025-04-30 13:03:20 -04:00
• 2ac74d098e [doc] add install tips (#17373) Reidandreidliu41 2025-05-01 01:02:41 +08:00
• 584f5fb4c6 [Bugfix][ROCm] Restrict ray version due to a breaking release (#17480) Gregory Shtrasberg 2025-04-30 12:59:06 -04:00
• d586ddc691 [BugFix] Fix authorization of openai_transcription_client.py (#17321) zh Wang 2025-05-01 00:51:05 +08:00
• 0b7e701dd4 [Docs] Update optimization.md doc (#17482) Michael Goin 2025-04-30 10:34:02 -06:00
• 947f2f5375 [V1] Allow turning off pickle fallback in vllm.v1.serial_utils (#17427) Russell BryantandCyrus Leung 2025-04-30 12:10:54 -04:00
• 739e03b344 [Bugfix] Fixed mistral tokenizer path when pointing to file (#17457) Pete Savage 2025-04-30 16:08:37 +01:00
• da4e7687b5 [Fix] Support passing args to logger (#17425) Aaron Pham 2025-04-30 11:06:58 -04:00
• 39317cf42b [Docs] Add command for running mypy tests from CI (#17475) Russell Bryant 2025-04-30 11:06:09 -04:00
• 2990cee95b [Feature] The Qwen3 reasoning parser supports guided decoding (#17466) Chauncey 2025-04-30 22:48:21 +08:00
• 0be6d05b5e [V1][Metrics] add support for kv event publishing (#16750) AlecandMark McLoughlin 2025-04-30 16:44:45 +02:00
• 77073c77bc [Core] Prevent side-channel attacks via cache salting (#17045) Marko Rosenmueller 2025-04-30 14:27:21 +02:00
• a7d5b016bd [TPU][V1][CI] Update regression test baseline for v6 CI (#17064) Nicolò Lucchesi 2025-04-30 13:03:22 +02:00
• d803786731 [V1][Bugfix]: vllm v1 verison metric num_gpu_blocks is None (#15755) rongfu.leng 2025-04-30 18:20:39 +08:00
• 1534d389af [Misc] Remove deprecated files (#17447) Chauncey 2025-04-30 16:52:19 +08:00
• ece5a8b0b6 Make the _apply_rotary_emb compatible with dynamo (#17435) Lu Fang 2025-04-30 00:52:48 -07:00
• 54072f315f [MODEL ADDITION] Ovis2 Model Addition (#15826) 3 people 2025-04-30 09:33:29 +02:00
• be633fba0f [Bugfix] Fix AttributeError: 'State' object has no attribute 'engine_client' (#17434) Chauncey 2025-04-30 15:11:04 +08:00
• ed6cfb90c8 [Hardware][Intel GPU] Upgrade to torch 2.7 (#17444) Kunshang JiandQiming Zhang 2025-04-30 15:03:58 +08:00
• 6ed9f6047e [Intel GPU] [CI]Fix XPU ci, setuptools >=80.0 have build issue (#17298) Kunshang Ji 2025-04-30 13:54:10 +08:00
• a44c4f1d2f Support LoRA for Mistral3 (#17428) Michael Goin 2025-04-29 22:10:30 -06:00
• 88fcf00dda Fix some speculative decode tests with tl.dot (#17371) Huy Do 2025-04-29 19:41:02 -07:00
• d1f569b1b9 Fix call to logger.info_once (#17416) Harry Mellor 2025-04-30 03:39:18 +01:00
• 13698db634 Improve configs - ModelConfig (#17130) Harry Mellor 2025-04-30 03:38:22 +01:00
• 2c4f59afc3 Update PyTorch to 2.7.0 (#16859) Huy Do 2025-04-29 19:08:04 -07:00
• 1c2bc7ead0 Truncation control for embedding models (#14776) Gabriel MarinhoandMax de Bayser 2025-04-29 22:24:57 -03:00
• 4055130a85 [release] Always git fetch all to get latest tag on TPU release (#17322) Kevin H. Luu 2025-04-29 17:52:11 -07:00
• 34120f5acd [V1][Feature] Enable Speculative Decoding with Structured Outputs (#14702) Benjamin Chislett 2025-04-29 17:02:10 -07:00
• 7489ec0bab Remove Bamba 9B from CI (#17407) Harry Mellor 2025-04-29 22:10:31 +01:00
• 70788bdbdc [V1][Spec Decode] Apply torch.compile & cudagraph to EAGLE (#17211) Bryan Lu 2025-04-29 14:10:00 -07:00
• c9c1b59e59 Fix: Python package installation for opentelmetry (#17049) Dilip Gowda Bhagavan 2025-04-30 01:50:24 +05:30
• 0350809f3a Remove Falcon3 2x7B from CI (#17404) Harry Mellor 2025-04-29 20:52:25 +01:00
• a6977dbd15 Simplify (and fix) passing of guided decoding backend options (#17008) Harry Mellor 2025-04-29 20:02:23 +01:00
• 2fa2a50bf9 [Bugfix] Fix Minicpm-O-int4 GPTQ model inference (#17397) Isotr0py 2025-04-30 02:21:42 +08:00
• 08e15defa9 [CI/Build] Add retry mechanism for add-apt-repository (#17107) Reidandreidliu41 2025-04-30 01:40:52 +08:00
• b37685afbb [CI] Uses Python 3.11 for TPU (#17359) Aaron Pham 2025-04-29 13:39:16 -04:00
• 792595b59d [TPU][V1][CI] Replace python3 setup.py develop with standard pip install --e on TPU (#17374) Nicolò Lucchesi 2025-04-29 19:36:48 +02:00
• 0c1c788312 [Doc][Typo] Fixing label in new model requests link in overview.md (#17400) casinca 2025-04-29 19:29:48 +02:00
• 56d64fbe30 [Docs] Propose a deprecation policy for the project (#17063) Russell BryantandCyrus Leung 2025-04-29 13:29:44 -04:00
• 608968b7c5 Enabling multi-group kernel tests. (#17115) Alexei-V-Ivanov-AMD 2025-04-29 12:27:27 -05:00