Commit Graph

Select branches

Hide Pull Requests

main

patches

sm86

Mono Color

• ad23318928 [Bugfix] Fixup Mamba (#10004) Tyler Michael Smith 2024-11-04 22:46:38 -05:00
• bbc3619dc8 [Core] Make encoder-decoder inputs a nested structure to be more composable (#9604) Cyrus Leung 2024-11-05 10:07:31 +08:00
• 04bbf38e05 [Core] Use os.sched_yield in ShmRingBuffer instead of time.sleep (#9994) Tyler Michael Smith 2024-11-04 20:08:21 -05:00
• 8f0a9ca890 [Bugfix] Respect modules_to_not_convert within awq_marlin (#9895) Michael Goin 2024-11-04 18:57:44 -05:00
• 2094062b4e [4.5/N] bugfix for quant config in speculative decode (#10007) youkaichao 2024-11-04 15:11:59 -08:00
• d93478b399 [Bugfix] Upgrade to pytorch 2.5.1 (#10001) bnellnm 2024-11-04 18:11:28 -05:00
• ac04a97a9f [Frontend] Add max_tokens prometheus metric (#9881) tomeras91 2024-11-05 00:53:24 +02:00
• 9a5664d4a4 [Misc] Refactor benchmark_throughput.py (#9779) 3 people 2024-11-04 14:32:16 -08:00
• 04cef2c6ab [Bugfix] Fix MQLLMEngine hanging (#9973) Robert Shaw 2024-11-04 16:01:43 -05:00
• 6e056bcf04 [Doc] Update VLM doc about loading from local files (#9999) Roger Wang 2024-11-04 11:47:11 -08:00
• 5208dc7a20 [Bugfix][CI/Build][Hardware][AMD] Shard ID parameters in AMD tests running parallel jobs (#9279) hissu-hyvarinen 2024-11-04 21:37:46 +02:00
• 1c45f4c385 [CI] Basic Integration Test For TPU (#9968) Robert Shaw 2024-11-04 14:34:26 -05:00
• 603a661ae8 [Model] factoring out MambaMixer out of Jamba (#8993) Mor Zusman 2024-11-04 20:00:00 +02:00
• fb2716d641 [Misc]Reduce BNB static variable (#9987) Jee Jee Li 2024-11-05 01:04:40 +08:00
• 8d72bb20fa [4/N] make quant config first-class citizen (#9978) youkaichao 2024-11-04 08:51:31 -08:00
• ac6b8f19b9 [Frontend] Multi-Modality Support for Loading Local Image Files (#9915) Chauncey 2024-11-04 23:34:57 +08:00
• ccb5376a9a [Bugfix][OpenVINO] Fix circular reference #9939 (#9974) Mengqing Cao 2024-11-04 18:14:13 +08:00
• ea4adeddc1 [Bugfix] Fix E2EL mean and median stats (#9984) Tran Quang Dai 2024-11-04 16:37:58 +07:00
• 4dbcbbeb09 [Misc] Compute query_start_loc/seq_start_loc on CPU (#9447) Yang ZhengandYang Zheng 2024-11-04 16:54:37 +08:00
• b67feb1274 [Bugfix]Using the correct type hints (#9885) Gregory Shtrasberg 2024-11-04 01:19:51 -05:00
• c49f0407ba [Bugfix] Fix MiniCPMV and Mllama BNB bug (#9917) Jee Jee Li 2024-11-04 11:36:41 +08:00
• 91c9ebbb1b [V1] Fix Configs (#9971) Robert Shaw 2024-11-03 19:24:40 -05:00
• 54597724f4 [Model] Add support for H2OVL-Mississippi models (#9747) shanshan wangandRoger Wang 2024-11-03 18:15:36 -06:00
• 1f1b6d6eda [V1] Support per-request seed (#9945) Nick Hill 2024-11-03 17:14:17 +00:00
• 3bb4befea7 [bugfix] fix tsts (#9959) youkaichao 2024-11-02 15:54:05 -07:00
• ae5279a163 [torch.compile] Adding torch compile to vision-language models (#9946) Yongzao 2024-11-03 03:56:05 +08:00
• 1b73ab2a1f [CI/Build] Quoting around > (#9956) Nikita Furin 2024-11-02 22:50:28 +03:00
• cea808f325 [3/N] model runner pass the whole config to model (#9958) youkaichao 2024-11-02 12:08:49 -07:00
• 74b529ceee [bugfix] fix chatglm dummy_data_for_glmv (#9955) youkaichao 2024-11-02 08:03:33 -07:00
• d6459b4516 [V1] Fix EngineArgs refactor on V1 (#9954) Robert Shaw 2024-11-02 10:44:38 -04:00
• e893795443 [2/N] executor pass the complete config to worker/modelrunner (#9938) youkaichaoandNick Hill 2024-11-02 07:35:05 -07:00
• 1d4cfe2be1 [Doc] Updated tpu-installation.rst with more details (#9926) Michael Green 2024-11-02 14:06:45 +00:00
• eed92f12fc [Docs] Update Granite 3.0 models in supported models table (#9930) Nick HillandCyrus Leung 2024-11-02 09:02:18 +00:00
• af7380d83b [torch.compile] fix cpu broken code (#9947) youkaichao 2024-11-01 23:35:47 -07:00
• a78dd3303e [Encoder Decoder] Add flash_attn kernel support for encoder-decoder models (#9559) sroy745 2024-11-01 23:22:49 -07:00
• d522034c85 [ci/build] Have dependabot ignore pinned dependencies (#9935) Kevin H. Luu 2024-11-01 13:56:13 -10:00
• 6c0b7f548d [Core][VLM] Add precise multi-modal placeholder tracking (#8346) Peter Salas 2024-11-01 16:21:10 -07:00
• d151fde834 [ci/build] Bump the patch-update group with 10 updates (#9897) 3 people 2024-11-01 23:04:42 +00:00
• 27cd36e6e2 [Bugfix] PicklingError on RayTaskError (#9934) Gene Der Su 2024-11-01 15:08:23 -07:00
• 18bd7587b7 [1/N] pass the complete config from engine to executor (#9933) youkaichao 2024-11-01 13:51:57 -07:00
• 598b6d7b07 [Bugfix/Core] Flashinfer k_scale and v_scale (#9861) Pavani Majety 2024-11-01 12:15:05 -07:00
• aff1fd8188 [torch.compile] use interpreter with stable api from pytorch (#9889) youkaichao 2024-11-01 11:50:37 -07:00
• 4581d2cc02 [Core] Refactor: Clean up unused argument in Scheduler._preempt (#9696) André Jonasson 2024-11-01 19:41:38 +01:00
• 1dd4cb2935 [Bugfix] Fix edge cases for MistralTokenizer (#9625) 3 people 2024-11-01 11:33:15 -06:00
• ba0d892074 [Frontend] Use a proper chat template for VLM2Vec (#9912) Cyrus Leung 2024-11-01 22:09:07 +08:00
• 30a2e80742 [CI/Build] Add Model Tests for PixtralHF (#9813) Michael Goin 2024-11-01 09:55:29 -04:00
• 06386a64dd [Frontend] Chat-based Embeddings API (#9759) Cyrus Leung 2024-11-01 16:13:35 +08:00
• d3aa2a8b2f [Doc] Update multi-input support (#9906) Cyrus Leung 2024-11-01 15:34:49 +08:00
• 2b5bf20988 [torch.compile] Adding torch compile annotations to some models (#9876) Yongzaoandyoukaichao 2024-11-01 15:25:47 +08:00
• 93a76dd21d [Model] Support bitsandbytes for MiniCPMV (#9891) Michael Goin 2024-11-01 01:31:56 -04:00
• 566cd27797 [torch.compile] rework test plans (#9866) youkaichao 2024-10-31 22:20:17 -07:00
• 37a4947dcd [Bugfix] Fix layer skip logic with bitsandbytes (#9887) Michael Goin 2024-11-01 01:12:44 -04:00
• 96e0c9cbbd [torch.compile] directly register custom op (#9896) youkaichao 2024-10-31 21:56:09 -07:00
• 031a7995f3 [Bugfix][Frontend] Reject guided decoding in multistep mode (#9892) Joe Runde 2024-10-31 19:09:46 -06:00
• b63c64d95b [ci/build] Configure dependabot to update pip dependencies (#9811) Kevin H. Luu 2024-10-31 12:55:38 -10:00
• 9fb12f7848 [BugFix][Kernel] Fix Illegal memory access in causal_conv1d in H100 (#9838) Mor Zusman 2024-10-31 22:06:25 +02:00
• 55650c83a0 [Bugfix] Fix illegal memory access error with chunked prefill, prefix caching, block manager v2 and xformers enabled together (#9532) sasha0552 2024-10-31 18:46:36 +00:00
• 77f7ef2908 [CI/Build] Adding a forced docker system prune to clean up space (#9849) Alexei-V-Ivanov-AMD 2024-10-31 12:02:58 -05:00
• 16b8f7a86f [CI/Build] Add Model Tests for Qwen2-VL (#9846) 3 people 2024-10-31 10:10:52 -06:00
• 5608e611c2 [Doc] Update Qwen documentation (#9869) Jee Jee Li 2024-10-31 16:54:18 +08:00
• 3ea2dc2ec4 [Misc] Remove deprecated arg for cuda graph capture (#9864) Roger Wang 2024-10-31 00:22:07 -07:00
• d087bf863e [Model] Support quantization of Qwen2VisionTransformer (#9817) Michael Goin 2024-10-31 01:41:20 -04:00
• 890ca36072 Revert "[Bugfix] Use host argument to bind to interface (#9798)" (#9852) Kevin H. Luu 2024-10-30 15:44:51 -10:00
• abbfb6134d [Misc][OpenAI] deprecate max_tokens in favor of new max_completion_tokens field for chat completion endpoint (#9837) Guillaume Calmettes 2024-10-31 02:15:56 +01:00
• 64384bbcdf [torch.compile] upgrade tests (#9858) youkaichao 2024-10-30 16:34:22 -07:00
• 00d91c8a2c [CI/Build] Simplify exception trace in api server tests (#9787) Yongzaoandyoukaichao 2024-10-31 05:52:05 +08:00
• c2cd1a2142 [doc] update pp support (#9853) youkaichao 2024-10-30 13:36:51 -07:00
• c787f2d81d [Neuron] Update Dockerfile.neuron to fix build failure (#9822) Harsha vardhan manoj Bikki 2024-10-30 12:22:02 -07:00
• 33d257735f [Doc] link bug for multistep guided decoding (#9843) Joe Runde 2024-10-30 11:28:29 -06:00
• 3b3f1e7436 [Bugfix][core] replace heartbeat with pid check (#9818) Joe Runde 2024-10-30 10:34:07 -06:00
• 9ff4511e43 [Misc] Add chunked-prefill support on FlashInfer. (#9781) Elfie Guo 2024-10-30 09:33:53 -07:00
• 81f09cfd80 [Model] Support math-shepherd-mistral-7b-prm model (#9697) Went-Liang 2024-10-31 00:33:42 +08:00
• cc98f1e079 [CI/Build] VLM Test Consolidation (#9372) Alex Brooks 2024-10-30 10:32:17 -06:00
• 211fe91aa8 [TPU] Correctly profile peak memory usage & Upgrade PyTorch XLA (#9438) Woosuk Kwon 2024-10-30 02:41:38 -07:00
• 6aa6020f9b [Misc] Specify minimum pynvml version (#9827) Jee Jee Li 2024-10-30 14:05:43 +08:00
• ff5ed6e1bc [torch.compile] rework compile control with piecewise cudagraph (#9715) youkaichao 2024-10-29 23:03:49 -07:00
• 7b0365efef [Doc] Add the DCO to CONTRIBUTING.md (#9803) 3 people 2024-10-30 01:22:23 -04:00
• 04a3ae0aca [Bugfix] Fix multi nodes TP+PP for XPU (#8884) Yan MaandYiSheng5 2024-10-30 12:34:45 +08:00
• 62fac4b9aa [ci/build] Pin CI dependencies version with pip-compile (#9810) Kevin H. Luu 2024-10-29 17:34:55 -10:00
• 226688bd61 [Bugfix][VLM] Make apply_fp8_linear work with >2D input (#9812) Michael Goin 2024-10-29 22:49:44 -04:00
• 64cb1cdc3f Update README.md (#9819) Lily Liu 2024-10-29 17:28:43 -07:00
• 1ab6f6b4ad [core][distributed] fix custom allreduce in pytorch 2.5 (#9815) youkaichao 2024-10-29 17:06:24 -07:00
• bc73e9821c [Bugfix] Fix prefix strings for quantized VLMs (#9772) Michael Goin 2024-10-29 19:02:59 -04:00
• 8d7724104a [Docs] Add notes about Snowflake Meetup (#9814) Simon Mo 2024-10-29 15:19:02 -07:00
• 882a1ad0de [Model] tool calling support for ibm-granite/granite-20b-functioncalling (#8339) 3 people 2024-10-29 18:07:37 -04:00
• 67bdf8e523 [Bugfix][Frontend] Guard against bad token ids (#9634) Joe Runde 2024-10-29 16:13:20 -05:00
• 0ad216f575 [MISC] Set label value to timestamp over 0, to keep track of recent history (#9777) Kunjan 2024-10-29 12:52:19 -07:00
• 7585ec996f [CI/Build] mergify: fix rules for ci/build label (#9804) Russell Bryant 2024-10-29 15:24:42 -04:00
• ab6f981671 [CI][Bugfix] Skip chameleon for transformers 4.46.1 (#9808) Michael Goin 2024-10-29 14:12:43 -04:00
• ac3d748dba [Model] Add LlamaEmbeddingModel as an embedding Implementation of LlamaModel (#9806) Junichi Sato 2024-10-30 02:40:35 +09:00
• 0ce7798f44 [Misc]: Typo fix: Renaming classes (casualLM -> causalLM) (#9801) yannicks1 2024-10-29 18:39:20 +01:00
• 0f43387157 [Bugfix] Use host argument to bind to interface (#9798) Sven Seeberg 2024-10-29 18:37:59 +01:00
• 08600ddc68 Fix the log to correct guide user to install modelscope (#9793) tastelikefeet 2024-10-30 01:36:59 +08:00
• 74fc2d77ae [Misc] Add metrics for request queue time, forward time, and execute time (#9659) 科英 2024-10-30 01:32:56 +08:00
• 622b7ab955 [Hardware] using current_platform.seed_everything (#9785) wangshuai09 2024-10-29 22:47:44 +08:00
• 09500f7dde [Model] Add BNB quantization support for Mllama (#9720) Isotr0py 2024-10-29 20:20:02 +08:00
• ef7865b4f9 [Frontend] re-enable multi-modality input in the new beam search implementation (#9427) Zhong Qishuai 2024-10-29 19:49:47 +08:00
• eae3d48181 [Bugfix] Use temporary directory in registry (#9721) Cyrus Leung 2024-10-29 13:08:20 +08:00
• e74f2d448c [Doc] Specify async engine args in docs (#9726) Cyrus Leung 2024-10-29 13:07:57 +08:00
• 7a4df5f200 [Model][LoRA]LoRA support added for Qwen (#9622) Jee Jee Li 2024-10-29 12:14:07 +08:00